{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9029495718363463,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_loss": NaN,
      "eval_runtime": 4.5948,
      "eval_samples_per_second": 4.788,
      "eval_steps_per_second": 4.788,
      "step": 0
    },
    {
      "epoch": 0.009514747859181731,
      "grad_norm": 0.0,
      "learning_rate": 1.9e-05,
      "loss": 2.0934,
      "step": 20
    },
    {
      "epoch": 0.019029495718363463,
      "grad_norm": 0.0,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 0.4738,
      "step": 40
    },
    {
      "epoch": 0.028544243577545196,
      "grad_norm": 0.0,
      "learning_rate": 5.9e-05,
      "loss": 0.8557,
      "step": 60
    },
    {
      "epoch": 0.038058991436726926,
      "grad_norm": 0.0,
      "learning_rate": 7.900000000000001e-05,
      "loss": 0.3978,
      "step": 80
    },
    {
      "epoch": 0.047573739295908656,
      "grad_norm": 0.0,
      "learning_rate": 9.900000000000001e-05,
      "loss": 0.4128,
      "step": 100
    },
    {
      "epoch": 0.05708848715509039,
      "grad_norm": 0.0,
      "learning_rate": 9.999471159635539e-05,
      "loss": 0.148,
      "step": 120
    },
    {
      "epoch": 0.06660323501427212,
      "grad_norm": 0.0,
      "learning_rate": 9.997771965008657e-05,
      "loss": 0.3554,
      "step": 140
    },
    {
      "epoch": 0.07611798287345385,
      "grad_norm": 0.0,
      "learning_rate": 9.994901349433758e-05,
      "loss": 1.0388,
      "step": 160
    },
    {
      "epoch": 0.08563273073263558,
      "grad_norm": 7.0538105964660645,
      "learning_rate": 9.990859985750506e-05,
      "loss": 0.6268,
      "step": 180
    },
    {
      "epoch": 0.09514747859181731,
      "grad_norm": 6.37161111831665,
      "learning_rate": 9.985648821208616e-05,
      "loss": 0.1791,
      "step": 200
    },
    {
      "epoch": 0.10466222645099905,
      "grad_norm": 0.0,
      "learning_rate": 9.979269077245831e-05,
      "loss": 0.0,
      "step": 220
    },
    {
      "epoch": 0.11417697431018078,
      "grad_norm": 0.0,
      "learning_rate": 9.97172224920163e-05,
      "loss": 0.4112,
      "step": 240
    },
    {
      "epoch": 0.12369172216936251,
      "grad_norm": 4.980894565582275,
      "learning_rate": 9.963010105966736e-05,
      "loss": 0.2527,
      "step": 260
    },
    {
      "epoch": 0.13320647002854424,
      "grad_norm": 0.0,
      "learning_rate": 9.953134689568506e-05,
      "loss": 0.639,
      "step": 280
    },
    {
      "epoch": 0.142721217887726,
      "grad_norm": 0.0,
      "learning_rate": 9.94209831469231e-05,
      "loss": 0.5911,
      "step": 300
    },
    {
      "epoch": 0.1522359657469077,
      "grad_norm": 0.0,
      "learning_rate": 9.929903568138989e-05,
      "loss": 0.0966,
      "step": 320
    },
    {
      "epoch": 0.16175071360608945,
      "grad_norm": 0.0,
      "learning_rate": 9.916553308218537e-05,
      "loss": 0.4412,
      "step": 340
    },
    {
      "epoch": 0.17126546146527116,
      "grad_norm": 0.0,
      "learning_rate": 9.902050664080152e-05,
      "loss": 0.3402,
      "step": 360
    },
    {
      "epoch": 0.1807802093244529,
      "grad_norm": 0.0,
      "learning_rate": 9.886399034978797e-05,
      "loss": 0.1149,
      "step": 380
    },
    {
      "epoch": 0.19029495718363462,
      "grad_norm": 0.0,
      "learning_rate": 9.869602089478447e-05,
      "loss": 0.1225,
      "step": 400
    },
    {
      "epoch": 0.19980970504281637,
      "grad_norm": 0.0,
      "learning_rate": 9.85166376459223e-05,
      "loss": 0.0718,
      "step": 420
    },
    {
      "epoch": 0.2093244529019981,
      "grad_norm": 0.0,
      "learning_rate": 9.832588264859624e-05,
      "loss": 0.3826,
      "step": 440
    },
    {
      "epoch": 0.21883920076117983,
      "grad_norm": 3.183824300765991,
      "learning_rate": 9.81238006136097e-05,
      "loss": 0.4913,
      "step": 460
    },
    {
      "epoch": 0.22835394862036157,
      "grad_norm": 0.0,
      "learning_rate": 9.7910438906695e-05,
      "loss": 0.0751,
      "step": 480
    },
    {
      "epoch": 0.23786869647954328,
      "grad_norm": 0.0,
      "learning_rate": 9.768584753741134e-05,
      "loss": 0.2444,
      "step": 500
    },
    {
      "epoch": 0.23786869647954328,
      "eval_loss": NaN,
      "eval_runtime": 8.8015,
      "eval_samples_per_second": 2.5,
      "eval_steps_per_second": 2.5,
      "step": 500
    },
    {
      "epoch": 0.24738344433872503,
      "grad_norm": 0.0,
      "learning_rate": 9.745007914742316e-05,
      "loss": 0.1088,
      "step": 520
    },
    {
      "epoch": 0.25689819219790677,
      "grad_norm": 0.0,
      "learning_rate": 9.720318899816155e-05,
      "loss": 0.1922,
      "step": 540
    },
    {
      "epoch": 0.2664129400570885,
      "grad_norm": 0.0,
      "learning_rate": 9.694523495787149e-05,
      "loss": 0.2443,
      "step": 560
    },
    {
      "epoch": 0.2759276879162702,
      "grad_norm": 0.0,
      "learning_rate": 9.667627748804835e-05,
      "loss": 0.0,
      "step": 580
    },
    {
      "epoch": 0.285442435775452,
      "grad_norm": 0.0,
      "learning_rate": 9.639637962926633e-05,
      "loss": 0.6168,
      "step": 600
    },
    {
      "epoch": 0.2949571836346337,
      "grad_norm": 0.0,
      "learning_rate": 9.610560698640241e-05,
      "loss": 0.249,
      "step": 620
    },
    {
      "epoch": 0.3044719314938154,
      "grad_norm": 0.0,
      "learning_rate": 9.580402771325941e-05,
      "loss": 0.1383,
      "step": 640
    },
    {
      "epoch": 0.3139866793529971,
      "grad_norm": 0.0,
      "learning_rate": 9.549171249659145e-05,
      "loss": 0.0658,
      "step": 660
    },
    {
      "epoch": 0.3235014272121789,
      "grad_norm": 0.0,
      "learning_rate": 9.51687345395358e-05,
      "loss": 0.7962,
      "step": 680
    },
    {
      "epoch": 0.3330161750713606,
      "grad_norm": 0.0,
      "learning_rate": 9.483516954445489e-05,
      "loss": 0.1979,
      "step": 700
    },
    {
      "epoch": 0.3425309229305423,
      "grad_norm": 0.0,
      "learning_rate": 9.449109569519253e-05,
      "loss": 0.0846,
      "step": 720
    },
    {
      "epoch": 0.3520456707897241,
      "grad_norm": 0.0,
      "learning_rate": 9.413659363874853e-05,
      "loss": 0.0,
      "step": 740
    },
    {
      "epoch": 0.3615604186489058,
      "grad_norm": 0.0,
      "learning_rate": 9.377174646637597e-05,
      "loss": 0.6378,
      "step": 760
    },
    {
      "epoch": 0.37107516650808753,
      "grad_norm": 0.0,
      "learning_rate": 9.339663969410546e-05,
      "loss": 0.1937,
      "step": 780
    },
    {
      "epoch": 0.38058991436726924,
      "grad_norm": 0.0,
      "learning_rate": 9.301136124270126e-05,
      "loss": 0.1475,
      "step": 800
    },
    {
      "epoch": 0.390104662226451,
      "grad_norm": 0.0,
      "learning_rate": 9.261600141705367e-05,
      "loss": 0.4155,
      "step": 820
    },
    {
      "epoch": 0.39961941008563273,
      "grad_norm": 0.0,
      "learning_rate": 9.221065288501245e-05,
      "loss": 0.4923,
      "step": 840
    },
    {
      "epoch": 0.40913415794481445,
      "grad_norm": 0.6096606254577637,
      "learning_rate": 9.179541065566664e-05,
      "loss": 0.1229,
      "step": 860
    },
    {
      "epoch": 0.4186489058039962,
      "grad_norm": 0.0,
      "learning_rate": 9.137037205707552e-05,
      "loss": 0.0,
      "step": 880
    },
    {
      "epoch": 0.42816365366317793,
      "grad_norm": 0.0,
      "learning_rate": 9.093563671345598e-05,
      "loss": 0.165,
      "step": 900
    },
    {
      "epoch": 0.43767840152235965,
      "grad_norm": 0.0,
      "learning_rate": 9.049130652183166e-05,
      "loss": 0.6362,
      "step": 920
    },
    {
      "epoch": 0.44719314938154137,
      "grad_norm": 0.0,
      "learning_rate": 9.003748562814962e-05,
      "loss": 0.0,
      "step": 940
    },
    {
      "epoch": 0.45670789724072314,
      "grad_norm": 0.0,
      "learning_rate": 8.957428040286959e-05,
      "loss": 0.0507,
      "step": 960
    },
    {
      "epoch": 0.46622264509990485,
      "grad_norm": 0.0,
      "learning_rate": 8.910179941603193e-05,
      "loss": 0.0257,
      "step": 980
    },
    {
      "epoch": 0.47573739295908657,
      "grad_norm": 4.7469587326049805,
      "learning_rate": 8.862015341181018e-05,
      "loss": 0.0655,
      "step": 1000
    },
    {
      "epoch": 0.47573739295908657,
      "eval_loss": NaN,
      "eval_runtime": 7.7821,
      "eval_samples_per_second": 2.827,
      "eval_steps_per_second": 2.827,
      "step": 1000
    },
    {
      "epoch": 0.48525214081826834,
      "grad_norm": 3.7392077445983887,
      "learning_rate": 8.812945528255368e-05,
      "loss": 0.4831,
      "step": 1020
    },
    {
      "epoch": 0.49476688867745006,
      "grad_norm": 0.0,
      "learning_rate": 8.762982004232709e-05,
      "loss": 0.0065,
      "step": 1040
    },
    {
      "epoch": 0.5042816365366318,
      "grad_norm": 0.0,
      "learning_rate": 8.712136479995226e-05,
      "loss": 0.0152,
      "step": 1060
    },
    {
      "epoch": 0.5137963843958135,
      "grad_norm": 0.0,
      "learning_rate": 8.66042087315593e-05,
      "loss": 0.079,
      "step": 1080
    },
    {
      "epoch": 0.5233111322549953,
      "grad_norm": 0.0,
      "learning_rate": 8.60784730526531e-05,
      "loss": 1.2652,
      "step": 1100
    },
    {
      "epoch": 0.532825880114177,
      "grad_norm": 0.0,
      "learning_rate": 8.554428098970171e-05,
      "loss": 0.0,
      "step": 1120
    },
    {
      "epoch": 0.5423406279733587,
      "grad_norm": 0.0,
      "learning_rate": 8.50017577512535e-05,
      "loss": 0.0428,
      "step": 1140
    },
    {
      "epoch": 0.5518553758325404,
      "grad_norm": 2.9529595375061035,
      "learning_rate": 8.445103049858966e-05,
      "loss": 0.3298,
      "step": 1160
    },
    {
      "epoch": 0.5613701236917221,
      "grad_norm": 0.0,
      "learning_rate": 8.389222831591914e-05,
      "loss": 0.0,
      "step": 1180
    },
    {
      "epoch": 0.570884871550904,
      "grad_norm": 0.0,
      "learning_rate": 8.332548218012263e-05,
      "loss": 0.0825,
      "step": 1200
    },
    {
      "epoch": 0.5803996194100857,
      "grad_norm": 0.0,
      "learning_rate": 8.275092493005321e-05,
      "loss": 0.2001,
      "step": 1220
    },
    {
      "epoch": 0.5899143672692674,
      "grad_norm": 0.0,
      "learning_rate": 8.216869123540034e-05,
      "loss": 0.6001,
      "step": 1240
    },
    {
      "epoch": 0.5994291151284491,
      "grad_norm": 1.827052116394043,
      "learning_rate": 8.157891756512488e-05,
      "loss": 0.0177,
      "step": 1260
    },
    {
      "epoch": 0.6089438629876308,
      "grad_norm": 0.0,
      "learning_rate": 8.098174215547224e-05,
      "loss": 0.0,
      "step": 1280
    },
    {
      "epoch": 0.6184586108468125,
      "grad_norm": 0.0,
      "learning_rate": 8.037730497757143e-05,
      "loss": 0.2685,
      "step": 1300
    },
    {
      "epoch": 0.6279733587059942,
      "grad_norm": 0.0,
      "learning_rate": 7.976574770462743e-05,
      "loss": 0.0258,
      "step": 1320
    },
    {
      "epoch": 0.6374881065651761,
      "grad_norm": 0.0,
      "learning_rate": 7.914721367871454e-05,
      "loss": 0.0001,
      "step": 1340
    },
    {
      "epoch": 0.6470028544243578,
      "grad_norm": 0.0,
      "learning_rate": 7.852184787717871e-05,
      "loss": 0.422,
      "step": 1360
    },
    {
      "epoch": 0.6565176022835395,
      "grad_norm": 3.8072144985198975,
      "learning_rate": 7.788979687865639e-05,
      "loss": 0.409,
      "step": 1380
    },
    {
      "epoch": 0.6660323501427212,
      "grad_norm": 0.0,
      "learning_rate": 7.725120882871828e-05,
      "loss": 0.4012,
      "step": 1400
    },
    {
      "epoch": 0.6755470980019029,
      "grad_norm": 0.0,
      "learning_rate": 7.660623340514561e-05,
      "loss": 0.0639,
      "step": 1420
    },
    {
      "epoch": 0.6850618458610847,
      "grad_norm": 0.0,
      "learning_rate": 7.595502178284743e-05,
      "loss": 0.1534,
      "step": 1440
    },
    {
      "epoch": 0.6945765937202664,
      "grad_norm": 0.0,
      "learning_rate": 7.529772659842685e-05,
      "loss": 0.0362,
      "step": 1460
    },
    {
      "epoch": 0.7040913415794482,
      "grad_norm": 0.0,
      "learning_rate": 7.46345019144048e-05,
      "loss": 0.0066,
      "step": 1480
    },
    {
      "epoch": 0.7136060894386299,
      "grad_norm": 0.0,
      "learning_rate": 7.396550318310957e-05,
      "loss": 0.7555,
      "step": 1500
    },
    {
      "epoch": 0.7136060894386299,
      "eval_loss": NaN,
      "eval_runtime": 7.7236,
      "eval_samples_per_second": 2.848,
      "eval_steps_per_second": 2.848,
      "step": 1500
    },
    {
      "epoch": 0.7231208372978116,
      "grad_norm": 1.2502919435501099,
      "learning_rate": 7.32908872102405e-05,
      "loss": 0.1364,
      "step": 1520
    },
    {
      "epoch": 0.7326355851569933,
      "grad_norm": 0.0,
      "learning_rate": 7.261081211811444e-05,
      "loss": 0.199,
      "step": 1540
    },
    {
      "epoch": 0.7421503330161751,
      "grad_norm": 0.0,
      "learning_rate": 7.192543730860388e-05,
      "loss": 0.2509,
      "step": 1560
    },
    {
      "epoch": 0.7516650808753568,
      "grad_norm": 0.0,
      "learning_rate": 7.123492342577472e-05,
      "loss": 0.1916,
      "step": 1580
    },
    {
      "epoch": 0.7611798287345385,
      "grad_norm": 0.0,
      "learning_rate": 7.053943231823328e-05,
      "loss": 0.3898,
      "step": 1600
    },
    {
      "epoch": 0.7706945765937203,
      "grad_norm": 3.7327072620391846,
      "learning_rate": 6.98391270011908e-05,
      "loss": 0.1748,
      "step": 1620
    },
    {
      "epoch": 0.780209324452902,
      "grad_norm": 0.0,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.1127,
      "step": 1640
    },
    {
      "epoch": 0.7897240723120837,
      "grad_norm": 0.0,
      "learning_rate": 6.842473140295417e-05,
      "loss": 0.0495,
      "step": 1660
    },
    {
      "epoch": 0.7992388201712655,
      "grad_norm": 0.0,
      "learning_rate": 6.77109726400134e-05,
      "loss": 0.1478,
      "step": 1680
    },
    {
      "epoch": 0.8087535680304472,
      "grad_norm": 0.0,
      "learning_rate": 6.69930626263742e-05,
      "loss": 0.5869,
      "step": 1700
    },
    {
      "epoch": 0.8182683158896289,
      "grad_norm": 0.0,
      "learning_rate": 6.627116963198458e-05,
      "loss": 0.1457,
      "step": 1720
    },
    {
      "epoch": 0.8277830637488106,
      "grad_norm": 0.0,
      "learning_rate": 6.554546286035802e-05,
      "loss": 0.671,
      "step": 1740
    },
    {
      "epoch": 0.8372978116079924,
      "grad_norm": 10.052992820739746,
      "learning_rate": 6.481611240891396e-05,
      "loss": 0.5565,
      "step": 1760
    },
    {
      "epoch": 0.8468125594671742,
      "grad_norm": 0.0,
      "learning_rate": 6.408328922910905e-05,
      "loss": 0.4085,
      "step": 1780
    },
    {
      "epoch": 0.8563273073263559,
      "grad_norm": 0.0,
      "learning_rate": 6.334716508636796e-05,
      "loss": 0.5145,
      "step": 1800
    },
    {
      "epoch": 0.8658420551855376,
      "grad_norm": 0.0,
      "learning_rate": 6.260791251982354e-05,
      "loss": 0.1931,
      "step": 1820
    },
    {
      "epoch": 0.8753568030447193,
      "grad_norm": 0.0,
      "learning_rate": 6.18657048018757e-05,
      "loss": 0.2708,
      "step": 1840
    },
    {
      "epoch": 0.884871550903901,
      "grad_norm": 0.0,
      "learning_rate": 6.112071589757814e-05,
      "loss": 0.3662,
      "step": 1860
    },
    {
      "epoch": 0.8943862987630827,
      "grad_norm": 0.0,
      "learning_rate": 6.037312042386314e-05,
      "loss": 0.0,
      "step": 1880
    },
    {
      "epoch": 0.9039010466222646,
      "grad_norm": 0.0,
      "learning_rate": 5.962309360861318e-05,
      "loss": 0.0005,
      "step": 1900
    },
    {
      "epoch": 0.9134157944814463,
      "grad_norm": 0.0,
      "learning_rate": 5.887081124958954e-05,
      "loss": 0.1641,
      "step": 1920
    },
    {
      "epoch": 0.922930542340628,
      "grad_norm": 0.0,
      "learning_rate": 5.811644967322737e-05,
      "loss": 0.118,
      "step": 1940
    },
    {
      "epoch": 0.9324452901998097,
      "grad_norm": 0.0,
      "learning_rate": 5.73601856933067e-05,
      "loss": 0.0,
      "step": 1960
    },
    {
      "epoch": 0.9419600380589914,
      "grad_norm": 0.0,
      "learning_rate": 5.660219656950937e-05,
      "loss": 0.0804,
      "step": 1980
    },
    {
      "epoch": 0.9514747859181731,
      "grad_norm": 0.0,
      "learning_rate": 5.584265996587129e-05,
      "loss": 0.3809,
      "step": 2000
    },
    {
      "epoch": 0.9514747859181731,
      "eval_loss": NaN,
      "eval_runtime": 7.6077,
      "eval_samples_per_second": 2.892,
      "eval_steps_per_second": 2.892,
      "step": 2000
    },
    {
      "epoch": 0.9609895337773549,
      "grad_norm": 0.0,
      "learning_rate": 5.5081753909140096e-05,
      "loss": 0.1075,
      "step": 2020
    },
    {
      "epoch": 0.9705042816365367,
      "grad_norm": 0.0,
      "learning_rate": 5.4319656747047654e-05,
      "loss": 0.5366,
      "step": 2040
    },
    {
      "epoch": 0.9800190294957184,
      "grad_norm": 0.0,
      "learning_rate": 5.355654710650737e-05,
      "loss": 0.3959,
      "step": 2060
    },
    {
      "epoch": 0.9895337773549001,
      "grad_norm": 0.0,
      "learning_rate": 5.2792603851746125e-05,
      "loss": 0.1957,
      "step": 2080
    },
    {
      "epoch": 0.9990485252140818,
      "grad_norm": 0.0,
      "learning_rate": 5.2028006042380474e-05,
      "loss": 0.184,
      "step": 2100
    },
    {
      "epoch": 1.0085632730732637,
      "grad_norm": 0.0,
      "learning_rate": 5.126293289144715e-05,
      "loss": 0.1857,
      "step": 2120
    },
    {
      "epoch": 1.0180780209324454,
      "grad_norm": 0.0,
      "learning_rate": 5.0497563723397526e-05,
      "loss": 0.5685,
      "step": 2140
    },
    {
      "epoch": 1.027592768791627,
      "grad_norm": 0.0,
      "learning_rate": 4.973207793206604e-05,
      "loss": 0.0807,
      "step": 2160
    },
    {
      "epoch": 1.0371075166508088,
      "grad_norm": 0.0,
      "learning_rate": 4.8966654938622295e-05,
      "loss": 0.6812,
      "step": 2180
    },
    {
      "epoch": 1.0466222645099905,
      "grad_norm": 0.0,
      "learning_rate": 4.820147414951677e-05,
      "loss": 0.0,
      "step": 2200
    },
    {
      "epoch": 1.0561370123691722,
      "grad_norm": 0.0,
      "learning_rate": 4.743671491443005e-05,
      "loss": 0.0253,
      "step": 2220
    },
    {
      "epoch": 1.065651760228354,
      "grad_norm": 2.1436240673065186,
      "learning_rate": 4.6672556484235174e-05,
      "loss": 0.1353,
      "step": 2240
    },
    {
      "epoch": 1.0751665080875357,
      "grad_norm": 0.0,
      "learning_rate": 4.590917796898342e-05,
      "loss": 0.2796,
      "step": 2260
    },
    {
      "epoch": 1.0846812559467174,
      "grad_norm": 0.0,
      "learning_rate": 4.514675829592278e-05,
      "loss": 0.1627,
      "step": 2280
    },
    {
      "epoch": 1.094196003805899,
      "grad_norm": 0.0,
      "learning_rate": 4.438547616755962e-05,
      "loss": 0.0606,
      "step": 2300
    },
    {
      "epoch": 1.1037107516650808,
      "grad_norm": 0.0,
      "learning_rate": 4.3625510019772715e-05,
      "loss": 0.008,
      "step": 2320
    },
    {
      "epoch": 1.1132254995242625,
      "grad_norm": 4.797143459320068,
      "learning_rate": 4.2867037979989957e-05,
      "loss": 0.1238,
      "step": 2340
    },
    {
      "epoch": 1.1227402473834442,
      "grad_norm": 0.0,
      "learning_rate": 4.211023782543727e-05,
      "loss": 0.1468,
      "step": 2360
    },
    {
      "epoch": 1.1322549952426262,
      "grad_norm": 0.0,
      "learning_rate": 4.135528694146974e-05,
      "loss": 0.4943,
      "step": 2380
    },
    {
      "epoch": 1.141769743101808,
      "grad_norm": 0.0,
      "learning_rate": 4.060236227999441e-05,
      "loss": 0.1699,
      "step": 2400
    },
    {
      "epoch": 1.1512844909609896,
      "grad_norm": 0.0,
      "learning_rate": 3.985164031799481e-05,
      "loss": 0.1025,
      "step": 2420
    },
    {
      "epoch": 1.1607992388201713,
      "grad_norm": 0.0,
      "learning_rate": 3.910329701616673e-05,
      "loss": 0.0501,
      "step": 2440
    },
    {
      "epoch": 1.170313986679353,
      "grad_norm": 0.0,
      "learning_rate": 3.8357507777675056e-05,
      "loss": 0.1253,
      "step": 2460
    },
    {
      "epoch": 1.1798287345385348,
      "grad_norm": 0.0,
      "learning_rate": 3.761444740704129e-05,
      "loss": 0.0181,
      "step": 2480
    },
    {
      "epoch": 1.1893434823977165,
      "grad_norm": 0.0,
      "learning_rate": 3.687429006917128e-05,
      "loss": 0.0237,
      "step": 2500
    },
    {
      "epoch": 1.1893434823977165,
      "eval_loss": NaN,
      "eval_runtime": 7.4307,
      "eval_samples_per_second": 2.961,
      "eval_steps_per_second": 2.961,
      "step": 2500
    },
    {
      "epoch": 1.1988582302568982,
      "grad_norm": 0.0,
      "learning_rate": 3.613720924853299e-05,
      "loss": 0.1829,
      "step": 2520
    },
    {
      "epoch": 1.20837297811608,
      "grad_norm": 0.0,
      "learning_rate": 3.540337770849371e-05,
      "loss": 0.0596,
      "step": 2540
    },
    {
      "epoch": 1.2178877259752616,
      "grad_norm": 0.0,
      "learning_rate": 3.467296745082634e-05,
      "loss": 0.0627,
      "step": 2560
    },
    {
      "epoch": 1.2274024738344433,
      "grad_norm": 0.0,
      "learning_rate": 3.394614967539402e-05,
      "loss": 0.0365,
      "step": 2580
    },
    {
      "epoch": 1.236917221693625,
      "grad_norm": 0.0,
      "learning_rate": 3.322309474002288e-05,
      "loss": 0.0,
      "step": 2600
    },
    {
      "epoch": 1.2464319695528068,
      "grad_norm": 0.0,
      "learning_rate": 3.25039721205721e-05,
      "loss": 0.0425,
      "step": 2620
    },
    {
      "epoch": 1.2559467174119887,
      "grad_norm": 0.0,
      "learning_rate": 3.178895037121077e-05,
      "loss": 0.0774,
      "step": 2640
    },
    {
      "epoch": 1.2654614652711702,
      "grad_norm": 0.0,
      "learning_rate": 3.107819708491059e-05,
      "loss": 0.5383,
      "step": 2660
    },
    {
      "epoch": 1.2749762131303521,
      "grad_norm": 0.0,
      "learning_rate": 3.037187885416398e-05,
      "loss": 0.0006,
      "step": 2680
    },
    {
      "epoch": 1.2844909609895339,
      "grad_norm": 0.0,
      "learning_rate": 2.967016123193682e-05,
      "loss": 0.0,
      "step": 2700
    },
    {
      "epoch": 1.2940057088487156,
      "grad_norm": 0.0,
      "learning_rate": 2.8973208692864624e-05,
      "loss": 0.0453,
      "step": 2720
    },
    {
      "epoch": 1.3035204567078973,
      "grad_norm": 0.0,
      "learning_rate": 2.828118459470156e-05,
      "loss": 0.0058,
      "step": 2740
    },
    {
      "epoch": 1.313035204567079,
      "grad_norm": 0.0,
      "learning_rate": 2.7594251140031223e-05,
      "loss": 0.0,
      "step": 2760
    },
    {
      "epoch": 1.3225499524262607,
      "grad_norm": 0.0,
      "learning_rate": 2.6912569338248315e-05,
      "loss": 0.0795,
      "step": 2780
    },
    {
      "epoch": 1.3320647002854424,
      "grad_norm": 0.0,
      "learning_rate": 2.623629896781969e-05,
      "loss": 0.1333,
      "step": 2800
    },
    {
      "epoch": 1.3415794481446242,
      "grad_norm": 0.0,
      "learning_rate": 2.5565598538834286e-05,
      "loss": 0.0725,
      "step": 2820
    },
    {
      "epoch": 1.3510941960038059,
      "grad_norm": 1.8207381963729858,
      "learning_rate": 2.4900625255849986e-05,
      "loss": 0.0143,
      "step": 2840
    },
    {
      "epoch": 1.3606089438629876,
      "grad_norm": 3.001304864883423,
      "learning_rate": 2.4241534981046815e-05,
      "loss": 0.0415,
      "step": 2860
    },
    {
      "epoch": 1.3701236917221693,
      "grad_norm": 2.418036699295044,
      "learning_rate": 2.3588482197694478e-05,
      "loss": 0.0665,
      "step": 2880
    },
    {
      "epoch": 1.379638439581351,
      "grad_norm": 0.0,
      "learning_rate": 2.294161997394336e-05,
      "loss": 0.1571,
      "step": 2900
    },
    {
      "epoch": 1.3891531874405327,
      "grad_norm": 0.0,
      "learning_rate": 2.2301099926946968e-05,
      "loss": 0.12,
      "step": 2920
    },
    {
      "epoch": 1.3986679352997147,
      "grad_norm": 0.0,
      "learning_rate": 2.1667072187324726e-05,
      "loss": 0.0335,
      "step": 2940
    },
    {
      "epoch": 1.4081826831588962,
      "grad_norm": 0.0,
      "learning_rate": 2.1039685363972934e-05,
      "loss": 0.0007,
      "step": 2960
    },
    {
      "epoch": 1.417697431018078,
      "grad_norm": 0.0,
      "learning_rate": 2.0419086509232648e-05,
      "loss": 0.0988,
      "step": 2980
    },
    {
      "epoch": 1.4272121788772598,
      "grad_norm": 0.0,
      "learning_rate": 1.9805421084422167e-05,
      "loss": 0.0749,
      "step": 3000
    },
    {
      "epoch": 1.4272121788772598,
      "eval_loss": NaN,
      "eval_runtime": 7.6652,
      "eval_samples_per_second": 2.87,
      "eval_steps_per_second": 2.87,
      "step": 3000
    },
    {
      "epoch": 1.4367269267364415,
      "grad_norm": 0.0,
      "learning_rate": 1.919883292574269e-05,
      "loss": 0.1049,
      "step": 3020
    },
    {
      "epoch": 1.4462416745956232,
      "grad_norm": 0.0,
      "learning_rate": 1.8599464210564586e-05,
      "loss": 0.0917,
      "step": 3040
    },
    {
      "epoch": 1.455756422454805,
      "grad_norm": 0.0,
      "learning_rate": 1.8007455424102797e-05,
      "loss": 0.3418,
      "step": 3060
    },
    {
      "epoch": 1.4652711703139867,
      "grad_norm": 0.0,
      "learning_rate": 1.7422945326488553e-05,
      "loss": 0.0139,
      "step": 3080
    },
    {
      "epoch": 1.4747859181731684,
      "grad_norm": 0.0,
      "learning_rate": 1.68460709202457e-05,
      "loss": 0.0101,
      "step": 3100
    },
    {
      "epoch": 1.4843006660323501,
      "grad_norm": 0.0,
      "learning_rate": 1.6276967418178746e-05,
      "loss": 0.0394,
      "step": 3120
    },
    {
      "epoch": 1.4938154138915318,
      "grad_norm": 0.0,
      "learning_rate": 1.5715768211680647e-05,
      "loss": 0.1327,
      "step": 3140
    },
    {
      "epoch": 1.5033301617507138,
      "grad_norm": 0.0,
      "learning_rate": 1.5162604839467265e-05,
      "loss": 0.0294,
      "step": 3160
    },
    {
      "epoch": 1.5128449096098953,
      "grad_norm": 0.0,
      "learning_rate": 1.4617606956746216e-05,
      "loss": 0.0,
      "step": 3180
    },
    {
      "epoch": 1.5223596574690772,
      "grad_norm": 0.0,
      "learning_rate": 1.4080902304827242e-05,
      "loss": 0.0,
      "step": 3200
    },
    {
      "epoch": 1.5318744053282587,
      "grad_norm": 0.0,
      "learning_rate": 1.3552616681181013e-05,
      "loss": 0.1559,
      "step": 3220
    },
    {
      "epoch": 1.5413891531874406,
      "grad_norm": 0.0,
      "learning_rate": 1.3032873909953636e-05,
      "loss": 0.0824,
      "step": 3240
    },
    {
      "epoch": 1.5509039010466221,
      "grad_norm": 0.0,
      "learning_rate": 1.2521795812943704e-05,
      "loss": 0.0029,
      "step": 3260
    },
    {
      "epoch": 1.560418648905804,
      "grad_norm": 0.0,
      "learning_rate": 1.2019502181048676e-05,
      "loss": 0.0081,
      "step": 3280
    },
    {
      "epoch": 1.5699333967649858,
      "grad_norm": 0.0,
      "learning_rate": 1.1526110746187224e-05,
      "loss": 0.1394,
      "step": 3300
    },
    {
      "epoch": 1.5794481446241675,
      "grad_norm": 0.0,
      "learning_rate": 1.1041737153704185e-05,
      "loss": 0.0001,
      "step": 3320
    },
    {
      "epoch": 1.5889628924833492,
      "grad_norm": 0.076429083943367,
      "learning_rate": 1.0566494935264625e-05,
      "loss": 0.0008,
      "step": 3340
    },
    {
      "epoch": 1.598477640342531,
      "grad_norm": 0.0,
      "learning_rate": 1.0100495482243356e-05,
      "loss": 0.0104,
      "step": 3360
    },
    {
      "epoch": 1.6079923882017126,
      "grad_norm": 0.0,
      "learning_rate": 9.643848019616003e-06,
      "loss": 0.0239,
      "step": 3380
    },
    {
      "epoch": 1.6175071360608944,
      "grad_norm": 0.0,
      "learning_rate": 9.196659580357913e-06,
      "loss": 0.0049,
      "step": 3400
    },
    {
      "epoch": 1.627021883920076,
      "grad_norm": 0.039900340139865875,
      "learning_rate": 8.759034980356862e-06,
      "loss": 0.0046,
      "step": 3420
    },
    {
      "epoch": 1.6365366317792578,
      "grad_norm": 0.0,
      "learning_rate": 8.33107679384542e-06,
      "loss": 0.2859,
      "step": 3440
    },
    {
      "epoch": 1.6460513796384397,
      "grad_norm": 0.0,
      "learning_rate": 7.912885329358688e-06,
      "loss": 0.193,
      "step": 3460
    },
    {
      "epoch": 1.6555661274976212,
      "grad_norm": 0.0,
      "learning_rate": 7.504558606223122e-06,
      "loss": 0.3163,
      "step": 3480
    },
    {
      "epoch": 1.6650808753568032,
      "grad_norm": 0.0,
      "learning_rate": 7.106192331581896e-06,
      "loss": 0.0896,
      "step": 3500
    },
    {
      "epoch": 1.6650808753568032,
      "eval_loss": NaN,
      "eval_runtime": 9.6486,
      "eval_samples_per_second": 2.28,
      "eval_steps_per_second": 2.28,
      "step": 3500
    },
    {
      "epoch": 1.6745956232159847,
      "grad_norm": 4.094954490661621,
      "learning_rate": 6.717879877962291e-06,
      "loss": 0.0605,
      "step": 3520
    },
    {
      "epoch": 1.6841103710751666,
      "grad_norm": 0.0,
      "learning_rate": 6.339712261390213e-06,
      "loss": 0.1816,
      "step": 3540
    },
    {
      "epoch": 1.693625118934348,
      "grad_norm": 0.0,
      "learning_rate": 5.971778120057031e-06,
      "loss": 0.0005,
      "step": 3560
    },
    {
      "epoch": 1.70313986679353,
      "grad_norm": 0.0,
      "learning_rate": 5.614163693543822e-06,
      "loss": 0.0328,
      "step": 3580
    },
    {
      "epoch": 1.7126546146527117,
      "grad_norm": 7.35809850692749,
      "learning_rate": 5.266952802607828e-06,
      "loss": 0.2526,
      "step": 3600
    },
    {
      "epoch": 1.7221693625118935,
      "grad_norm": 0.0,
      "learning_rate": 4.930226829535767e-06,
      "loss": 0.1301,
      "step": 3620
    },
    {
      "epoch": 1.7316841103710752,
      "grad_norm": 0.0,
      "learning_rate": 4.604064699068766e-06,
      "loss": 0.0367,
      "step": 3640
    },
    {
      "epoch": 1.7411988582302569,
      "grad_norm": 0.0,
      "learning_rate": 4.288542859903316e-06,
      "loss": 0.0089,
      "step": 3660
    },
    {
      "epoch": 1.7507136060894386,
      "grad_norm": 0.0,
      "learning_rate": 3.983735266772565e-06,
      "loss": 0.0477,
      "step": 3680
    },
    {
      "epoch": 1.7602283539486203,
      "grad_norm": 0.0,
      "learning_rate": 3.689713363112146e-06,
      "loss": 0.0,
      "step": 3700
    },
    {
      "epoch": 1.7697431018078023,
      "grad_norm": 0.0,
      "learning_rate": 3.406546064314664e-06,
      "loss": 0.0074,
      "step": 3720
    },
    {
      "epoch": 1.7792578496669837,
      "grad_norm": 2.901078462600708,
      "learning_rate": 3.1342997415767015e-06,
      "loss": 0.0307,
      "step": 3740
    },
    {
      "epoch": 1.7887725975261657,
      "grad_norm": 0.0,
      "learning_rate": 2.873038206342188e-06,
      "loss": 0.0974,
      "step": 3760
    },
    {
      "epoch": 1.7982873453853472,
      "grad_norm": 0.0,
      "learning_rate": 2.622822695345706e-06,
      "loss": 0.0024,
      "step": 3780
    },
    {
      "epoch": 1.8078020932445291,
      "grad_norm": 0.0,
      "learning_rate": 2.3837118562592797e-06,
      "loss": 0.0944,
      "step": 3800
    },
    {
      "epoch": 1.8173168411037106,
      "grad_norm": 0.0,
      "learning_rate": 2.1557617339460432e-06,
      "loss": 0.0281,
      "step": 3820
    },
    {
      "epoch": 1.8268315889628925,
      "grad_norm": 0.0,
      "learning_rate": 1.939025757323987e-06,
      "loss": 0.1706,
      "step": 3840
    },
    {
      "epoch": 1.8363463368220743,
      "grad_norm": 3.4238052368164062,
      "learning_rate": 1.7335547268427843e-06,
      "loss": 0.0456,
      "step": 3860
    },
    {
      "epoch": 1.845861084681256,
      "grad_norm": 0.0,
      "learning_rate": 1.5393968025767702e-06,
      "loss": 0.2662,
      "step": 3880
    },
    {
      "epoch": 1.8553758325404377,
      "grad_norm": 0.0,
      "learning_rate": 1.3565974929367397e-06,
      "loss": 0.1209,
      "step": 3900
    },
    {
      "epoch": 1.8648905803996194,
      "grad_norm": 0.0,
      "learning_rate": 1.1851996440033319e-06,
      "loss": 0.1809,
      "step": 3920
    },
    {
      "epoch": 1.8744053282588011,
      "grad_norm": 0.0,
      "learning_rate": 1.0252434294843737e-06,
      "loss": 0.1648,
      "step": 3940
    },
    {
      "epoch": 1.8839200761179828,
      "grad_norm": 0.0,
      "learning_rate": 8.767663412986127e-07,
      "loss": 0.3638,
      "step": 3960
    },
    {
      "epoch": 1.8934348239771646,
      "grad_norm": 0.0,
      "learning_rate": 7.398031807880457e-07,
      "loss": 0.0223,
      "step": 3980
    },
    {
      "epoch": 1.9029495718363463,
      "grad_norm": 0.0,
      "learning_rate": 6.143860505608945e-07,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 1.9029495718363463,
      "eval_loss": NaN,
      "eval_runtime": 8.4613,
      "eval_samples_per_second": 2.6,
      "eval_steps_per_second": 2.6,
      "step": 4000
    }
  ],
  "logging_steps": 20,
  "max_steps": 4204,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.658384411983872e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}