{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 625,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0016,
      "grad_norm": 4.720448017120361,
      "learning_rate": 3.1746031746031746e-06,
      "loss": 1.8429,
      "mean_token_accuracy": 0.6849687695503235,
      "step": 1
    },
    {
      "epoch": 0.008,
      "grad_norm": 4.610220909118652,
      "learning_rate": 1.5873015873015872e-05,
      "loss": 1.8484,
      "mean_token_accuracy": 0.6779065430164337,
      "step": 5
    },
    {
      "epoch": 0.016,
      "grad_norm": 2.8285515308380127,
      "learning_rate": 3.1746031746031745e-05,
      "loss": 1.7723,
      "mean_token_accuracy": 0.6799486577510834,
      "step": 10
    },
    {
      "epoch": 0.024,
      "grad_norm": 1.520603895187378,
      "learning_rate": 4.761904761904762e-05,
      "loss": 1.5141,
      "mean_token_accuracy": 0.7008779108524322,
      "step": 15
    },
    {
      "epoch": 0.032,
      "grad_norm": 1.0638386011123657,
      "learning_rate": 6.349206349206349e-05,
      "loss": 1.2605,
      "mean_token_accuracy": 0.7299223065376281,
      "step": 20
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9480394721031189,
      "learning_rate": 7.936507936507937e-05,
      "loss": 0.9797,
      "mean_token_accuracy": 0.7793598055839539,
      "step": 25
    },
    {
      "epoch": 0.048,
      "grad_norm": 0.8817314505577087,
      "learning_rate": 9.523809523809524e-05,
      "loss": 0.6795,
      "mean_token_accuracy": 0.8407030582427979,
      "step": 30
    },
    {
      "epoch": 0.056,
      "grad_norm": 0.8544878363609314,
      "learning_rate": 0.00011111111111111112,
      "loss": 0.3682,
      "mean_token_accuracy": 0.9016113102436065,
      "step": 35
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.5889844298362732,
      "learning_rate": 0.00012698412698412698,
      "loss": 0.1517,
      "mean_token_accuracy": 0.9596078157424927,
      "step": 40
    },
    {
      "epoch": 0.072,
      "grad_norm": 0.2203538715839386,
      "learning_rate": 0.00014285714285714287,
      "loss": 0.0955,
      "mean_token_accuracy": 0.9677912712097168,
      "step": 45
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.1825166940689087,
      "learning_rate": 0.00015873015873015873,
      "loss": 0.0826,
      "mean_token_accuracy": 0.9692689836025238,
      "step": 50
    },
    {
      "epoch": 0.088,
      "grad_norm": 0.15790687501430511,
      "learning_rate": 0.00017460317460317462,
      "loss": 0.0851,
      "mean_token_accuracy": 0.9700896739959717,
      "step": 55
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.1395537108182907,
      "learning_rate": 0.00019047619047619048,
      "loss": 0.0786,
      "mean_token_accuracy": 0.9702458798885345,
      "step": 60
    },
    {
      "epoch": 0.104,
      "grad_norm": 0.14210280776023865,
      "learning_rate": 0.00019999375039475277,
      "loss": 0.0774,
      "mean_token_accuracy": 0.9697415053844451,
      "step": 65
    },
    {
      "epoch": 0.112,
      "grad_norm": 0.12832307815551758,
      "learning_rate": 0.0001999234513064475,
      "loss": 0.0734,
      "mean_token_accuracy": 0.9710267901420593,
      "step": 70
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.10479996353387833,
      "learning_rate": 0.00019977509622105233,
      "loss": 0.0745,
      "mean_token_accuracy": 0.9724782824516296,
      "step": 75
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.1173420250415802,
      "learning_rate": 0.0001995488010273198,
      "loss": 0.0729,
      "mean_token_accuracy": 0.9719331085681915,
      "step": 80
    },
    {
      "epoch": 0.136,
      "grad_norm": 0.08444946259260178,
      "learning_rate": 0.00019924474249753655,
      "loss": 0.0695,
      "mean_token_accuracy": 0.9733094036579132,
      "step": 85
    },
    {
      "epoch": 0.144,
      "grad_norm": 0.13057316839694977,
      "learning_rate": 0.00019886315814943647,
      "loss": 0.0739,
      "mean_token_accuracy": 0.9714005470275879,
      "step": 90
    },
    {
      "epoch": 0.152,
      "grad_norm": 0.2025141566991806,
      "learning_rate": 0.0001984043460606618,
      "loss": 0.0736,
      "mean_token_accuracy": 0.9701524496078491,
      "step": 95
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1009383425116539,
      "learning_rate": 0.0001978686646359173,
      "loss": 0.0728,
      "mean_token_accuracy": 0.9720687448978425,
      "step": 100
    },
    {
      "epoch": 0.168,
      "grad_norm": 0.10014893114566803,
      "learning_rate": 0.0001972565323269996,
      "loss": 0.0715,
      "mean_token_accuracy": 0.9728178203105926,
      "step": 105
    },
    {
      "epoch": 0.176,
      "grad_norm": 0.116126149892807,
      "learning_rate": 0.00019656842730592046,
      "loss": 0.0711,
      "mean_token_accuracy": 0.9717484176158905,
      "step": 110
    },
    {
      "epoch": 0.184,
      "grad_norm": 0.08928077667951584,
      "learning_rate": 0.0001958048870913786,
      "loss": 0.0682,
      "mean_token_accuracy": 0.974213844537735,
      "step": 115
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.11367285251617432,
      "learning_rate": 0.0001949665081288729,
      "loss": 0.0706,
      "mean_token_accuracy": 0.9731513261795044,
      "step": 120
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.08238652348518372,
      "learning_rate": 0.00019405394532478424,
      "loss": 0.0696,
      "mean_token_accuracy": 0.9734784185886383,
      "step": 125
    },
    {
      "epoch": 0.208,
      "grad_norm": 0.11063515394926071,
      "learning_rate": 0.00019306791153479006,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9721537351608276,
      "step": 130
    },
    {
      "epoch": 0.216,
      "grad_norm": 0.15030214190483093,
      "learning_rate": 0.00019200917700701176,
      "loss": 0.0696,
      "mean_token_accuracy": 0.9727127254009247,
      "step": 135
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.07141675055027008,
      "learning_rate": 0.0001908785687803289,
      "loss": 0.0698,
      "mean_token_accuracy": 0.9731925427913666,
      "step": 140
    },
    {
      "epoch": 0.232,
      "grad_norm": 0.07116198539733887,
      "learning_rate": 0.00018967697003833157,
      "loss": 0.0683,
      "mean_token_accuracy": 0.9736765503883362,
      "step": 145
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.08784201741218567,
      "learning_rate": 0.0001884053194194142,
      "loss": 0.0703,
      "mean_token_accuracy": 0.9728135228157043,
      "step": 150
    },
    {
      "epoch": 0.248,
      "grad_norm": 0.09302135556936264,
      "learning_rate": 0.00018706461028355104,
      "loss": 0.0722,
      "mean_token_accuracy": 0.9723609626293183,
      "step": 155
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.10910028219223022,
      "learning_rate": 0.00018565588993632487,
      "loss": 0.0705,
      "mean_token_accuracy": 0.9722101032733917,
      "step": 160
    },
    {
      "epoch": 0.264,
      "grad_norm": 0.04893859848380089,
      "learning_rate": 0.0001841802588108161,
      "loss": 0.0738,
      "mean_token_accuracy": 0.971207445859909,
      "step": 165
    },
    {
      "epoch": 0.272,
      "grad_norm": 0.12638282775878906,
      "learning_rate": 0.00018263886960799062,
      "loss": 0.0712,
      "mean_token_accuracy": 0.973378598690033,
      "step": 170
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.056649304926395416,
      "learning_rate": 0.00018103292639625837,
      "loss": 0.069,
      "mean_token_accuracy": 0.9724430739879608,
      "step": 175
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.12598562240600586,
      "learning_rate": 0.0001793636836709057,
      "loss": 0.0717,
      "mean_token_accuracy": 0.9714934885501861,
      "step": 180
    },
    {
      "epoch": 0.296,
      "grad_norm": 0.07508666068315506,
      "learning_rate": 0.0001776324453741365,
      "loss": 0.072,
      "mean_token_accuracy": 0.9719970643520355,
      "step": 185
    },
    {
      "epoch": 0.304,
      "grad_norm": 0.07222873717546463,
      "learning_rate": 0.00017584056387648727,
      "loss": 0.0685,
      "mean_token_accuracy": 0.9727974116802216,
      "step": 190
    },
    {
      "epoch": 0.312,
      "grad_norm": 0.05881664529442787,
      "learning_rate": 0.0001739894389204122,
      "loss": 0.0719,
      "mean_token_accuracy": 0.9726415634155273,
      "step": 195
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.11033733189105988,
      "learning_rate": 0.00017208051652686335,
      "loss": 0.0702,
      "mean_token_accuracy": 0.97275031208992,
      "step": 200
    },
    {
      "epoch": 0.328,
      "grad_norm": 0.10474064946174622,
      "learning_rate": 0.00017011528786571969,
      "loss": 0.0689,
      "mean_token_accuracy": 0.974085009098053,
      "step": 205
    },
    {
      "epoch": 0.336,
      "grad_norm": 0.0579833947122097,
      "learning_rate": 0.00016809528809094807,
      "loss": 0.0671,
      "mean_token_accuracy": 0.9740655779838562,
      "step": 210
    },
    {
      "epoch": 0.344,
      "grad_norm": 0.1026485487818718,
      "learning_rate": 0.0001660220951414055,
      "loss": 0.0695,
      "mean_token_accuracy": 0.9734614491462708,
      "step": 215
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.07079461216926575,
      "learning_rate": 0.00016389732850821966,
      "loss": 0.0664,
      "mean_token_accuracy": 0.9740577757358551,
      "step": 220
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.07856956869363785,
      "learning_rate": 0.0001617226479697105,
      "loss": 0.0698,
      "mean_token_accuracy": 0.973381620645523,
      "step": 225
    },
    {
      "epoch": 0.368,
      "grad_norm": 0.08537988364696503,
      "learning_rate": 0.00015949975229484134,
      "loss": 0.0687,
      "mean_token_accuracy": 0.9732668340206146,
      "step": 230
    },
    {
      "epoch": 0.376,
      "grad_norm": 0.07466263324022293,
      "learning_rate": 0.00015723037791621193,
      "loss": 0.0695,
      "mean_token_accuracy": 0.9729018092155457,
      "step": 235
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.08313712477684021,
      "learning_rate": 0.00015491629757363032,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9735385298728942,
      "step": 240
    },
    {
      "epoch": 0.392,
      "grad_norm": 0.04696296900510788,
      "learning_rate": 0.00015255931892932333,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9733852624893189,
      "step": 245
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.06557007133960724,
      "learning_rate": 0.0001501612831558664,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9734211206436157,
      "step": 250
    },
    {
      "epoch": 0.408,
      "grad_norm": 0.11457318812608719,
      "learning_rate": 0.00014772406349793744,
      "loss": 0.0692,
      "mean_token_accuracy": 0.9734779357910156,
      "step": 255
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.09975489228963852,
      "learning_rate": 0.0001452495638090167,
      "loss": 0.0696,
      "mean_token_accuracy": 0.9729597091674804,
      "step": 260
    },
    {
      "epoch": 0.424,
      "grad_norm": 0.0677042007446289,
      "learning_rate": 0.00014273971706417647,
      "loss": 0.0691,
      "mean_token_accuracy": 0.9730812609195709,
      "step": 265
    },
    {
      "epoch": 0.432,
      "grad_norm": 0.054182957857847214,
      "learning_rate": 0.00014019648385012244,
      "loss": 0.069,
      "mean_token_accuracy": 0.972840279340744,
      "step": 270
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.07722967118024826,
      "learning_rate": 0.00013762185083366556,
      "loss": 0.0678,
      "mean_token_accuracy": 0.9736016094684601,
      "step": 275
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.07115447521209717,
      "learning_rate": 0.00013501782920982184,
      "loss": 0.0691,
      "mean_token_accuracy": 0.972719419002533,
      "step": 280
    },
    {
      "epoch": 0.456,
      "grad_norm": 0.09139351546764374,
      "learning_rate": 0.00013238645313075104,
      "loss": 0.0694,
      "mean_token_accuracy": 0.9720758616924285,
      "step": 285
    },
    {
      "epoch": 0.464,
      "grad_norm": 0.09019674360752106,
      "learning_rate": 0.00012972977811676287,
      "loss": 0.0674,
      "mean_token_accuracy": 0.974070566892624,
      "step": 290
    },
    {
      "epoch": 0.472,
      "grad_norm": 0.05537855625152588,
      "learning_rate": 0.00012704987945063068,
      "loss": 0.0745,
      "mean_token_accuracy": 0.972056758403778,
      "step": 295
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.08490055799484253,
      "learning_rate": 0.00012434885055646823,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9741802930831909,
      "step": 300
    },
    {
      "epoch": 0.488,
      "grad_norm": 0.06208372861146927,
      "learning_rate": 0.00012162880136443447,
      "loss": 0.0672,
      "mean_token_accuracy": 0.9741501808166504,
      "step": 305
    },
    {
      "epoch": 0.496,
      "grad_norm": 0.06884725391864777,
      "learning_rate": 0.00011889185666254506,
      "loss": 0.0683,
      "mean_token_accuracy": 0.9735860764980316,
      "step": 310
    },
    {
      "epoch": 0.504,
      "grad_norm": 0.0937989354133606,
      "learning_rate": 0.00011614015443687722,
      "loss": 0.068,
      "mean_token_accuracy": 0.973924046754837,
      "step": 315
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.05234411731362343,
      "learning_rate": 0.0001133758442014651,
      "loss": 0.0675,
      "mean_token_accuracy": 0.9733489453792572,
      "step": 320
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.06664064526557922,
      "learning_rate": 0.00011060108531918971,
      "loss": 0.0677,
      "mean_token_accuracy": 0.9736957907676697,
      "step": 325
    },
    {
      "epoch": 0.528,
      "grad_norm": 0.055740658193826675,
      "learning_rate": 0.0001078180453149754,
      "loss": 0.0675,
      "mean_token_accuracy": 0.973907732963562,
      "step": 330
    },
    {
      "epoch": 0.536,
      "grad_norm": 0.05514289066195488,
      "learning_rate": 0.00010502889818261075,
      "loss": 0.0675,
      "mean_token_accuracy": 0.974098140001297,
      "step": 335
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.08672691881656647,
      "learning_rate": 0.00010223582268651586,
      "loss": 0.0677,
      "mean_token_accuracy": 0.9738192439079285,
      "step": 340
    },
    {
      "epoch": 0.552,
      "grad_norm": 0.07068340480327606,
      "learning_rate": 9.94410006597835e-05,
      "loss": 0.0677,
      "mean_token_accuracy": 0.9725731074810028,
      "step": 345
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.09604080766439438,
      "learning_rate": 9.66466152998226e-05,
      "loss": 0.067,
      "mean_token_accuracy": 0.9743229866027832,
      "step": 350
    },
    {
      "epoch": 0.568,
      "grad_norm": 0.07730761170387268,
      "learning_rate": 9.385484946293637e-05,
      "loss": 0.0669,
      "mean_token_accuracy": 0.9742228567600251,
      "step": 355
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.05239711329340935,
      "learning_rate": 9.106788395916678e-05,
      "loss": 0.0672,
      "mean_token_accuracy": 0.9740143895149231,
      "step": 360
    },
    {
      "epoch": 0.584,
      "grad_norm": 0.06987932324409485,
      "learning_rate": 8.828789584873754e-05,
      "loss": 0.0667,
      "mean_token_accuracy": 0.9739366114139557,
      "step": 365
    },
    {
      "epoch": 0.592,
      "grad_norm": 0.07997996360063553,
      "learning_rate": 8.551705674142617e-05,
      "loss": 0.0669,
      "mean_token_accuracy": 0.9741333663463593,
      "step": 370
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.08489294350147247,
      "learning_rate": 8.275753110019367e-05,
      "loss": 0.0668,
      "mean_token_accuracy": 0.9730526447296143,
      "step": 375
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.07549357414245605,
      "learning_rate": 8.001147455039735e-05,
      "loss": 0.0659,
      "mean_token_accuracy": 0.9744420766830444,
      "step": 380
    },
    {
      "epoch": 0.616,
      "grad_norm": 0.06531010568141937,
      "learning_rate": 7.728103219590681e-05,
      "loss": 0.0659,
      "mean_token_accuracy": 0.9741898834705353,
      "step": 385
    },
    {
      "epoch": 0.624,
      "grad_norm": 0.0814213976264,
      "learning_rate": 7.456833694343906e-05,
      "loss": 0.068,
      "mean_token_accuracy": 0.9731835901737214,
      "step": 390
    },
    {
      "epoch": 0.632,
      "grad_norm": 0.06650282442569733,
      "learning_rate": 7.18755078364214e-05,
      "loss": 0.0664,
      "mean_token_accuracy": 0.9736463844776153,
      "step": 395
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.10044140368700027,
      "learning_rate": 6.920464839968405e-05,
      "loss": 0.0676,
      "mean_token_accuracy": 0.9736545443534851,
      "step": 400
    },
    {
      "epoch": 0.648,
      "grad_norm": 0.05335594713687897,
      "learning_rate": 6.65578449962749e-05,
      "loss": 0.0668,
      "mean_token_accuracy": 0.9744662940502167,
      "step": 405
    },
    {
      "epoch": 0.656,
      "grad_norm": 0.06943535059690475,
      "learning_rate": 6.393716519768047e-05,
      "loss": 0.0673,
      "mean_token_accuracy": 0.9734912216663361,
      "step": 410
    },
    {
      "epoch": 0.664,
      "grad_norm": 0.060714706778526306,
      "learning_rate": 6.134465616872598e-05,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9740535855293274,
      "step": 415
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.2205849438905716,
      "learning_rate": 5.878234306841637e-05,
      "loss": 0.0685,
      "mean_token_accuracy": 0.9733618140220642,
      "step": 420
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.07674142718315125,
      "learning_rate": 5.62522274679673e-05,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9741897404193878,
      "step": 425
    },
    {
      "epoch": 0.688,
      "grad_norm": 0.08134385943412781,
      "learning_rate": 5.375628578726181e-05,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9741408348083496,
      "step": 430
    },
    {
      "epoch": 0.696,
      "grad_norm": 0.07496479898691177,
      "learning_rate": 5.1296467750954314e-05,
      "loss": 0.0674,
      "mean_token_accuracy": 0.9741273105144501,
      "step": 435
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.061479806900024414,
      "learning_rate": 4.8874694865427676e-05,
      "loss": 0.0669,
      "mean_token_accuracy": 0.973869401216507,
      "step": 440
    },
    {
      "epoch": 0.712,
      "grad_norm": 0.0543564110994339,
      "learning_rate": 4.649285891779327e-05,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9746919453144074,
      "step": 445
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.07584322243928909,
      "learning_rate": 4.415282049810644e-05,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9739439964294434,
      "step": 450
    },
    {
      "epoch": 0.728,
      "grad_norm": 0.04671603813767433,
      "learning_rate": 4.1856407545951834e-05,
      "loss": 0.067,
      "mean_token_accuracy": 0.973672068119049,
      "step": 455
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.06221921369433403,
      "learning_rate": 3.9605413922533874e-05,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9742021799087525,
      "step": 460
    },
    {
      "epoch": 0.744,
      "grad_norm": 0.045539796352386475,
      "learning_rate": 3.740159800938784e-05,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9738292276859284,
      "step": 465
    },
    {
      "epoch": 0.752,
      "grad_norm": 0.0595044270157814,
      "learning_rate": 3.5246681334806175e-05,
      "loss": 0.0669,
      "mean_token_accuracy": 0.9747107863426209,
      "step": 470
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.06676959246397018,
      "learning_rate": 3.3142347229053015e-05,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9743025422096252,
      "step": 475
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.05794461444020271,
      "learning_rate": 3.109023950941736e-05,
      "loss": 0.0651,
      "mean_token_accuracy": 0.9745893716812134,
      "step": 480
    },
    {
      "epoch": 0.776,
      "grad_norm": 0.06010904908180237,
      "learning_rate": 2.909196119613218e-05,
      "loss": 0.0663,
      "mean_token_accuracy": 0.9746502339839935,
      "step": 485
    },
    {
      "epoch": 0.784,
      "grad_norm": 0.05273396894335747,
      "learning_rate": 2.7149073260162416e-05,
      "loss": 0.0657,
      "mean_token_accuracy": 0.9738177537918091,
      "step": 490
    },
    {
      "epoch": 0.792,
      "grad_norm": 0.07098868489265442,
      "learning_rate": 2.5263093403840142e-05,
      "loss": 0.0657,
      "mean_token_accuracy": 0.9750109136104583,
      "step": 495
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.07257623225450516,
      "learning_rate": 2.3435494875299314e-05,
      "loss": 0.0651,
      "mean_token_accuracy": 0.9739606738090515,
      "step": 500
    },
    {
      "epoch": 0.808,
      "grad_norm": 0.09082420915365219,
      "learning_rate": 2.166770531763633e-05,
      "loss": 0.0672,
      "mean_token_accuracy": 0.9737493813037872,
      "step": 505
    },
    {
      "epoch": 0.816,
      "grad_norm": 0.08171214908361435,
      "learning_rate": 1.9961105653695266e-05,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9747599720954895,
      "step": 510
    },
    {
      "epoch": 0.824,
      "grad_norm": 0.0967075377702713,
      "learning_rate": 1.8317029007349085e-05,
      "loss": 0.0664,
      "mean_token_accuracy": 0.9733656942844391,
      "step": 515
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.04340047016739845,
      "learning_rate": 1.6736759662119183e-05,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9745591640472412,
      "step": 520
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.04629400745034218,
      "learning_rate": 1.5221532057947419e-05,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9746206283569336,
      "step": 525
    },
    {
      "epoch": 0.848,
      "grad_norm": 0.0683581680059433,
      "learning_rate": 1.3772529826903269e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.9746199965476989,
      "step": 530
    },
    {
      "epoch": 0.856,
      "grad_norm": 0.06316008418798447,
      "learning_rate": 1.23908848685804e-05,
      "loss": 0.0663,
      "mean_token_accuracy": 0.9744003236293792,
      "step": 535
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.06182578578591347,
      "learning_rate": 1.1077676465904208e-05,
      "loss": 0.0675,
      "mean_token_accuracy": 0.9744610726833344,
      "step": 540
    },
    {
      "epoch": 0.872,
      "grad_norm": 0.07484375685453415,
      "learning_rate": 9.833930442041506e-06,
      "loss": 0.067,
      "mean_token_accuracy": 0.9738114476203918,
      "step": 545
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.05372326448559761,
      "learning_rate": 8.660618359070604e-06,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9736744463443756,
      "step": 550
    },
    {
      "epoch": 0.888,
      "grad_norm": 0.0624370239675045,
      "learning_rate": 7.558656759037797e-06,
      "loss": 0.0668,
      "mean_token_accuracy": 0.9736717522144318,
      "step": 555
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.06488735973834991,
      "learning_rate": 6.528906447993288e-06,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9738507032394409,
      "step": 560
    },
    {
      "epoch": 0.904,
      "grad_norm": 0.06491252779960632,
      "learning_rate": 5.572171823565797e-06,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9741439461708069,
      "step": 565
    },
    {
      "epoch": 0.912,
      "grad_norm": 0.06388004124164581,
      "learning_rate": 4.689200246600867e-06,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9750186562538147,
      "step": 570
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.09783890843391418,
      "learning_rate": 3.880681457354118e-06,
      "loss": 0.067,
      "mean_token_accuracy": 0.9736691117286682,
      "step": 575
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.05274941399693489,
      "learning_rate": 3.1472470366950334e-06,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9749404191970825,
      "step": 580
    },
    {
      "epoch": 0.936,
      "grad_norm": 0.05393306538462639,
      "learning_rate": 2.4894699127426367e-06,
      "loss": 0.0672,
      "mean_token_accuracy": 0.9738192737102509,
      "step": 585
    },
    {
      "epoch": 0.944,
      "grad_norm": 0.057169899344444275,
      "learning_rate": 1.907863913318153e-06,
      "loss": 0.0661,
      "mean_token_accuracy": 0.974576759338379,
      "step": 590
    },
    {
      "epoch": 0.952,
      "grad_norm": 0.06199713051319122,
      "learning_rate": 1.4028833645643113e-06,
      "loss": 0.0671,
      "mean_token_accuracy": 0.9740838885307312,
      "step": 595
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.053632188588380814,
      "learning_rate": 9.749227360448143e-07,
      "loss": 0.0657,
      "mean_token_accuracy": 0.9742229998111724,
      "step": 600
    },
    {
      "epoch": 0.968,
      "grad_norm": 0.0578572042286396,
      "learning_rate": 6.243163326014267e-07,
      "loss": 0.0648,
      "mean_token_accuracy": 0.9743983149528503,
      "step": 605
    },
    {
      "epoch": 0.976,
      "grad_norm": 0.07826316356658936,
      "learning_rate": 3.5133803320896994e-07,
      "loss": 0.0671,
      "mean_token_accuracy": 0.9743796646595001,
      "step": 610
    },
    {
      "epoch": 0.984,
      "grad_norm": 0.2104119062423706,
      "learning_rate": 1.562010770326916e-07,
      "loss": 0.0666,
      "mean_token_accuracy": 0.97383713722229,
      "step": 615
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.04197349026799202,
      "learning_rate": 3.905789685471062e-08,
      "loss": 0.0653,
      "mean_token_accuracy": 0.9747089684009552,
      "step": 620
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.07076921314001083,
      "learning_rate": 0.0,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9736450254917145,
      "step": 625
    },
    {
      "epoch": 1.0,
      "step": 625,
      "total_flos": 4738039144775680.0,
      "train_loss": 0.13301027479171754,
      "train_runtime": 396.9652,
      "train_samples_per_second": 12.596,
      "train_steps_per_second": 1.574
    }
  ],
  "logging_steps": 5,
  "max_steps": 625,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4738039144775680.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}