{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 659,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.3487173642963171,
      "epoch": 0.015186028853454821,
      "grad_norm": 1.609375,
      "learning_rate": 2.7272727272727272e-06,
      "loss": 0.4099268913269043,
      "mean_token_accuracy": 0.8717762351036071,
      "num_tokens": 568708.0,
      "step": 10
    },
    {
      "entropy": 0.37818768359720706,
      "epoch": 0.030372057706909643,
      "grad_norm": 0.87890625,
      "learning_rate": 5.7575757575757586e-06,
      "loss": 0.39949469566345214,
      "mean_token_accuracy": 0.8737476468086243,
      "num_tokens": 1125639.0,
      "step": 20
    },
    {
      "entropy": 0.40618006475269797,
      "epoch": 0.04555808656036447,
      "grad_norm": 0.70703125,
      "learning_rate": 8.787878787878788e-06,
      "loss": 0.3975033760070801,
      "mean_token_accuracy": 0.8727035835385323,
      "num_tokens": 1683225.0,
      "step": 30
    },
    {
      "entropy": 0.3896496780216694,
      "epoch": 0.060744115413819286,
      "grad_norm": 0.6953125,
      "learning_rate": 9.997733473639876e-06,
      "loss": 0.3925030708312988,
      "mean_token_accuracy": 0.8742863699793816,
      "num_tokens": 2236895.0,
      "step": 40
    },
    {
      "entropy": 0.37322904225438835,
      "epoch": 0.07593014426727411,
      "grad_norm": 0.68359375,
      "learning_rate": 9.983889919973586e-06,
      "loss": 0.3752753257751465,
      "mean_token_accuracy": 0.8792506881058216,
      "num_tokens": 2818707.0,
      "step": 50
    },
    {
      "entropy": 0.3811411205679178,
      "epoch": 0.09111617312072894,
      "grad_norm": 0.66796875,
      "learning_rate": 9.957496810072027e-06,
      "loss": 0.38604438304901123,
      "mean_token_accuracy": 0.8750339619815349,
      "num_tokens": 3351348.0,
      "step": 60
    },
    {
      "entropy": 0.3796327030286193,
      "epoch": 0.10630220197418375,
      "grad_norm": 0.66015625,
      "learning_rate": 9.918620602428916e-06,
      "loss": 0.37710745334625245,
      "mean_token_accuracy": 0.8776259452104569,
      "num_tokens": 3915545.0,
      "step": 70
    },
    {
      "entropy": 0.37812459245324137,
      "epoch": 0.12148823082763857,
      "grad_norm": 0.64453125,
      "learning_rate": 9.867359188282193e-06,
      "loss": 0.38009963035583494,
      "mean_token_accuracy": 0.8783061921596527,
      "num_tokens": 4462906.0,
      "step": 80
    },
    {
      "entropy": 0.3751340739428997,
      "epoch": 0.1366742596810934,
      "grad_norm": 0.6640625,
      "learning_rate": 9.803841645121505e-06,
      "loss": 0.37636594772338866,
      "mean_token_accuracy": 0.8778362341225148,
      "num_tokens": 5029003.0,
      "step": 90
    },
    {
      "entropy": 0.37551863975822924,
      "epoch": 0.15186028853454822,
      "grad_norm": 0.6640625,
      "learning_rate": 9.728227911667934e-06,
      "loss": 0.3773549795150757,
      "mean_token_accuracy": 0.8772883579134941,
      "num_tokens": 5596042.0,
      "step": 100
    },
    {
      "entropy": 0.3809764288365841,
      "epoch": 0.16704631738800305,
      "grad_norm": 0.71484375,
      "learning_rate": 9.640708385144403e-06,
      "loss": 0.3807323932647705,
      "mean_token_accuracy": 0.8774459846317768,
      "num_tokens": 6144821.0,
      "step": 110
    },
    {
      "entropy": 0.37467240951955316,
      "epoch": 0.18223234624145787,
      "grad_norm": 0.62890625,
      "learning_rate": 9.541503441850844e-06,
      "loss": 0.37542564868927003,
      "mean_token_accuracy": 0.8782215595245362,
      "num_tokens": 6691491.0,
      "step": 120
    },
    {
      "entropy": 0.3787713166326284,
      "epoch": 0.19741837509491267,
      "grad_norm": 0.7109375,
      "learning_rate": 9.430862882251279e-06,
      "loss": 0.37993783950805665,
      "mean_token_accuracy": 0.8774278596043587,
      "num_tokens": 7247335.0,
      "step": 130
    },
    {
      "entropy": 0.3862619888037443,
      "epoch": 0.2126044039483675,
      "grad_norm": 0.71484375,
      "learning_rate": 9.309065301970193e-06,
      "loss": 0.38727219104766847,
      "mean_token_accuracy": 0.8749251998960972,
      "num_tokens": 7808664.0,
      "step": 140
    },
    {
      "entropy": 0.3778634283691645,
      "epoch": 0.22779043280182232,
      "grad_norm": 0.71484375,
      "learning_rate": 9.176417390281944e-06,
      "loss": 0.38028583526611326,
      "mean_token_accuracy": 0.8772468723356723,
      "num_tokens": 8360893.0,
      "step": 150
    },
    {
      "entropy": 0.3749677825719118,
      "epoch": 0.24297646165527714,
      "grad_norm": 0.69921875,
      "learning_rate": 9.033253157859715e-06,
      "loss": 0.37344467639923096,
      "mean_token_accuracy": 0.8786589197814465,
      "num_tokens": 8905139.0,
      "step": 160
    },
    {
      "entropy": 0.37994367331266404,
      "epoch": 0.25816249050873197,
      "grad_norm": 0.69140625,
      "learning_rate": 8.879933095728485e-06,
      "loss": 0.38379650115966796,
      "mean_token_accuracy": 0.8768095754086971,
      "num_tokens": 9467791.0,
      "step": 170
    },
    {
      "entropy": 0.3774459037929773,
      "epoch": 0.2733485193621868,
      "grad_norm": 0.7109375,
      "learning_rate": 8.716843267539868e-06,
      "loss": 0.3767258644104004,
      "mean_token_accuracy": 0.8779186218976974,
      "num_tokens": 10013526.0,
      "step": 180
    },
    {
      "entropy": 0.3706828704103827,
      "epoch": 0.2885345482156416,
      "grad_norm": 0.6875,
      "learning_rate": 8.544394337454409e-06,
      "loss": 0.373125958442688,
      "mean_token_accuracy": 0.8792334951460361,
      "num_tokens": 10567209.0,
      "step": 190
    },
    {
      "entropy": 0.3781122103333473,
      "epoch": 0.30372057706909644,
      "grad_norm": 0.7109375,
      "learning_rate": 8.36302053607924e-06,
      "loss": 0.3779691457748413,
      "mean_token_accuracy": 0.877835976332426,
      "num_tokens": 11121802.0,
      "step": 200
    },
    {
      "entropy": 0.37852676026523113,
      "epoch": 0.31890660592255127,
      "grad_norm": 0.703125,
      "learning_rate": 8.17317856706482e-06,
      "loss": 0.37905910015106203,
      "mean_token_accuracy": 0.877592646330595,
      "num_tokens": 11677885.0,
      "step": 210
    },
    {
      "entropy": 0.376834512129426,
      "epoch": 0.3340926347760061,
      "grad_norm": 0.66015625,
      "learning_rate": 7.975346457114034e-06,
      "loss": 0.3753563404083252,
      "mean_token_accuracy": 0.8776590585708618,
      "num_tokens": 12235216.0,
      "step": 220
    },
    {
      "entropy": 0.3737819105386734,
      "epoch": 0.3492786636294609,
      "grad_norm": 0.671875,
      "learning_rate": 7.770022352299294e-06,
      "loss": 0.37358593940734863,
      "mean_token_accuracy": 0.878921328485012,
      "num_tokens": 12787759.0,
      "step": 230
    },
    {
      "entropy": 0.3769740372896194,
      "epoch": 0.36446469248291574,
      "grad_norm": 0.73046875,
      "learning_rate": 7.557723263718596e-06,
      "loss": 0.37995898723602295,
      "mean_token_accuracy": 0.8769361607730388,
      "num_tokens": 13346471.0,
      "step": 240
    },
    {
      "entropy": 0.3878506176173687,
      "epoch": 0.37965072133637057,
      "grad_norm": 0.70703125,
      "learning_rate": 7.338983765648985e-06,
      "loss": 0.38782215118408203,
      "mean_token_accuracy": 0.8749015353620052,
      "num_tokens": 13895194.0,
      "step": 250
    },
    {
      "entropy": 0.37555828876793385,
      "epoch": 0.39483675018982534,
      "grad_norm": 0.6875,
      "learning_rate": 7.114354649475499e-06,
      "loss": 0.3771331787109375,
      "mean_token_accuracy": 0.878202386945486,
      "num_tokens": 14453542.0,
      "step": 260
    },
    {
      "entropy": 0.37179951313883064,
      "epoch": 0.41002277904328016,
      "grad_norm": 0.6484375,
      "learning_rate": 6.884401536785045e-06,
      "loss": 0.37206058502197265,
      "mean_token_accuracy": 0.8789021499454975,
      "num_tokens": 15016280.0,
      "step": 270
    },
    {
      "entropy": 0.3706459369510412,
      "epoch": 0.425208807896735,
      "grad_norm": 0.7109375,
      "learning_rate": 6.6497034551174585e-06,
      "loss": 0.37101426124572756,
      "mean_token_accuracy": 0.8798882246017456,
      "num_tokens": 15561057.0,
      "step": 280
    },
    {
      "entropy": 0.37728526555001735,
      "epoch": 0.4403948367501898,
      "grad_norm": 0.65625,
      "learning_rate": 6.41085137996006e-06,
      "loss": 0.37785754203796384,
      "mean_token_accuracy": 0.8779311388731003,
      "num_tokens": 16127699.0,
      "step": 290
    },
    {
      "entropy": 0.3816283464431763,
      "epoch": 0.45558086560364464,
      "grad_norm": 0.81640625,
      "learning_rate": 6.168446746656973e-06,
      "loss": 0.3794879674911499,
      "mean_token_accuracy": 0.8773063771426678,
      "num_tokens": 16686457.0,
      "step": 300
    },
    {
      "entropy": 0.37517447732388975,
      "epoch": 0.47076689445709946,
      "grad_norm": 0.70703125,
      "learning_rate": 5.923099935980278e-06,
      "loss": 0.3782352924346924,
      "mean_token_accuracy": 0.8787827685475349,
      "num_tokens": 17254272.0,
      "step": 310
    },
    {
      "entropy": 0.374018133059144,
      "epoch": 0.4859529233105543,
      "grad_norm": 0.71484375,
      "learning_rate": 5.675428737176367e-06,
      "loss": 0.37341156005859377,
      "mean_token_accuracy": 0.8788688823580741,
      "num_tokens": 17809900.0,
      "step": 320
    },
    {
      "entropy": 0.3753270395100117,
      "epoch": 0.5011389521640092,
      "grad_norm": 0.68359375,
      "learning_rate": 5.426056792357552e-06,
      "loss": 0.3752497673034668,
      "mean_token_accuracy": 0.8784179173409938,
      "num_tokens": 18379566.0,
      "step": 330
    },
    {
      "entropy": 0.3742110010236502,
      "epoch": 0.5163249810174639,
      "grad_norm": 0.6875,
      "learning_rate": 5.175612026156045e-06,
      "loss": 0.3746063232421875,
      "mean_token_accuracy": 0.8782069273293018,
      "num_tokens": 18943281.0,
      "step": 340
    },
    {
      "entropy": 0.37444472052156924,
      "epoch": 0.5315110098709187,
      "grad_norm": 0.71484375,
      "learning_rate": 4.924725064594448e-06,
      "loss": 0.3729024171829224,
      "mean_token_accuracy": 0.8787923693656922,
      "num_tokens": 19488865.0,
      "step": 350
    },
    {
      "entropy": 0.3750518877059221,
      "epoch": 0.5466970387243736,
      "grad_norm": 0.78515625,
      "learning_rate": 4.674027647154037e-06,
      "loss": 0.3758077621459961,
      "mean_token_accuracy": 0.8765743866562843,
      "num_tokens": 20048281.0,
      "step": 360
    },
    {
      "entropy": 0.3787516813725233,
      "epoch": 0.5618830675778284,
      "grad_norm": 0.74609375,
      "learning_rate": 4.424151036039381e-06,
      "loss": 0.3790909767150879,
      "mean_token_accuracy": 0.8769759923219681,
      "num_tokens": 20597434.0,
      "step": 370
    },
    {
      "entropy": 0.3786877432838082,
      "epoch": 0.5770690964312832,
      "grad_norm": 0.68359375,
      "learning_rate": 4.175724426644724e-06,
      "loss": 0.3812232971191406,
      "mean_token_accuracy": 0.8777030549943448,
      "num_tokens": 21161267.0,
      "step": 380
    },
    {
      "entropy": 0.37311795353889465,
      "epoch": 0.592255125284738,
      "grad_norm": 0.6796875,
      "learning_rate": 3.929373363224654e-06,
      "loss": 0.3731100559234619,
      "mean_token_accuracy": 0.8793233536183834,
      "num_tokens": 21709421.0,
      "step": 390
    },
    {
      "entropy": 0.3734915753826499,
      "epoch": 0.6074411541381929,
      "grad_norm": 0.6875,
      "learning_rate": 3.685718163758427e-06,
      "loss": 0.37124335765838623,
      "mean_token_accuracy": 0.8786324210464954,
      "num_tokens": 22250023.0,
      "step": 400
    },
    {
      "entropy": 0.3722789943218231,
      "epoch": 0.6226271829916477,
      "grad_norm": 0.67578125,
      "learning_rate": 3.445372357974194e-06,
      "loss": 0.37429609298706057,
      "mean_token_accuracy": 0.8784996062517166,
      "num_tokens": 22802881.0,
      "step": 410
    },
    {
      "entropy": 0.3826644644141197,
      "epoch": 0.6378132118451025,
      "grad_norm": 0.65625,
      "learning_rate": 3.2089411424661864e-06,
      "loss": 0.3828511953353882,
      "mean_token_accuracy": 0.875868634134531,
      "num_tokens": 23368508.0,
      "step": 420
    },
    {
      "entropy": 0.36304581388831136,
      "epoch": 0.6529992406985573,
      "grad_norm": 0.703125,
      "learning_rate": 2.977019856794955e-06,
      "loss": 0.362534499168396,
      "mean_token_accuracy": 0.8821237675845623,
      "num_tokens": 23923709.0,
      "step": 430
    },
    {
      "entropy": 0.38759873658418653,
      "epoch": 0.6681852695520122,
      "grad_norm": 0.67578125,
      "learning_rate": 2.7501924844078538e-06,
      "loss": 0.38718571662902834,
      "mean_token_accuracy": 0.8746685221791267,
      "num_tokens": 24477925.0,
      "step": 440
    },
    {
      "entropy": 0.3708066754043102,
      "epoch": 0.683371298405467,
      "grad_norm": 0.6875,
      "learning_rate": 2.5290301821544826e-06,
      "loss": 0.36970815658569334,
      "mean_token_accuracy": 0.8802599847316742,
      "num_tokens": 25027245.0,
      "step": 450
    },
    {
      "entropy": 0.36688214987516404,
      "epoch": 0.6985573272589218,
      "grad_norm": 0.68359375,
      "learning_rate": 2.3140898420998425e-06,
      "loss": 0.3657586097717285,
      "mean_token_accuracy": 0.8809232294559479,
      "num_tokens": 25582534.0,
      "step": 460
    },
    {
      "entropy": 0.36983687337487936,
      "epoch": 0.7137433561123766,
      "grad_norm": 0.7109375,
      "learning_rate": 2.105912689256533e-06,
      "loss": 0.37239837646484375,
      "mean_token_accuracy": 0.8794391065835953,
      "num_tokens": 26134875.0,
      "step": 470
    },
    {
      "entropy": 0.3732773784548044,
      "epoch": 0.7289293849658315,
      "grad_norm": 0.63671875,
      "learning_rate": 1.905022918766995e-06,
      "loss": 0.37306258678436277,
      "mean_token_accuracy": 0.8793606124818325,
      "num_tokens": 26681084.0,
      "step": 480
    },
    {
      "entropy": 0.3710829207673669,
      "epoch": 0.7441154138192863,
      "grad_norm": 0.69921875,
      "learning_rate": 1.7119263759673677e-06,
      "loss": 0.3711911678314209,
      "mean_token_accuracy": 0.8803343921899796,
      "num_tokens": 27234332.0,
      "step": 490
    },
    {
      "entropy": 0.3832434043288231,
      "epoch": 0.7593014426727411,
      "grad_norm": 0.9921875,
      "learning_rate": 1.5271092826566108e-06,
      "loss": 0.3841698169708252,
      "mean_token_accuracy": 0.8759766638278961,
      "num_tokens": 27794602.0,
      "step": 500
    },
    {
      "entropy": 0.38211295008659363,
      "epoch": 0.7744874715261959,
      "grad_norm": 0.8046875,
      "learning_rate": 1.3510370127781635e-06,
      "loss": 0.3804590940475464,
      "mean_token_accuracy": 0.8769169762730599,
      "num_tokens": 28354831.0,
      "step": 510
    },
    {
      "entropy": 0.3752284612506628,
      "epoch": 0.7896735003796507,
      "grad_norm": 0.73046875,
      "learning_rate": 1.1841529205970281e-06,
      "loss": 0.37546916007995607,
      "mean_token_accuracy": 0.8786770381033421,
      "num_tokens": 28922852.0,
      "step": 520
    },
    {
      "entropy": 0.36873019095510245,
      "epoch": 0.8048595292331056,
      "grad_norm": 0.6640625,
      "learning_rate": 1.026877224322923e-06,
      "loss": 0.36797375679016114,
      "mean_token_accuracy": 0.880731363594532,
      "num_tokens": 29493442.0,
      "step": 530
    },
    {
      "entropy": 0.3788988694548607,
      "epoch": 0.8200455580865603,
      "grad_norm": 0.69921875,
      "learning_rate": 8.7960594799059e-07,
      "loss": 0.37884984016418455,
      "mean_token_accuracy": 0.8770281121134758,
      "num_tokens": 30034443.0,
      "step": 540
    },
    {
      "entropy": 0.3814096964895725,
      "epoch": 0.8352315869400152,
      "grad_norm": 0.73046875,
      "learning_rate": 7.427099242616348e-07,
      "loss": 0.3821078300476074,
      "mean_token_accuracy": 0.8763406798243523,
      "num_tokens": 30570567.0,
      "step": 550
    },
    {
      "entropy": 0.3758743409067392,
      "epoch": 0.85041761579347,
      "grad_norm": 0.6953125,
      "learning_rate": 6.165338606588517e-07,
      "loss": 0.3744307279586792,
      "mean_token_accuracy": 0.8794760994613171,
      "num_tokens": 31129457.0,
      "step": 560
    },
    {
      "entropy": 0.37431446108967065,
      "epoch": 0.8656036446469249,
      "grad_norm": 0.6640625,
      "learning_rate": 5.0139547158427e-07,
      "loss": 0.37335963249206544,
      "mean_token_accuracy": 0.8793582506477833,
      "num_tokens": 31689902.0,
      "step": 570
    },
    {
      "entropy": 0.38372623883187773,
      "epoch": 0.8807896735003796,
      "grad_norm": 0.7578125,
      "learning_rate": 3.9758467830656623e-07,
      "loss": 0.38321547508239745,
      "mean_token_accuracy": 0.8755873307585716,
      "num_tokens": 32253359.0,
      "step": 580
    },
    {
      "entropy": 0.36962624490261076,
      "epoch": 0.8959757023538345,
      "grad_norm": 0.72265625,
      "learning_rate": 3.0536287893223603e-07,
      "loss": 0.37100839614868164,
      "mean_token_accuracy": 0.8799250744283199,
      "num_tokens": 32813428.0,
      "step": 590
    },
    {
      "entropy": 0.3853254303336143,
      "epoch": 0.9111617312072893,
      "grad_norm": 0.6640625,
      "learning_rate": 2.2496229019879635e-07,
      "loss": 0.3848439693450928,
      "mean_token_accuracy": 0.8762004837393761,
      "num_tokens": 33382024.0,
      "step": 600
    },
    {
      "entropy": 0.3742272950708866,
      "epoch": 0.9263477600607442,
      "grad_norm": 0.671875,
      "learning_rate": 1.5658536274738623e-07,
      "loss": 0.3725078582763672,
      "mean_token_accuracy": 0.8787065915763378,
      "num_tokens": 33939521.0,
      "step": 610
    },
    {
      "entropy": 0.3774934906512499,
      "epoch": 0.9415337889141989,
      "grad_norm": 0.69140625,
      "learning_rate": 1.004042713471165e-07,
      "loss": 0.37858588695526124,
      "mean_token_accuracy": 0.877676124125719,
      "num_tokens": 34482539.0,
      "step": 620
    },
    {
      "entropy": 0.37360552567988636,
      "epoch": 0.9567198177676538,
      "grad_norm": 0.734375,
      "learning_rate": 5.6560481354807625e-08,
      "loss": 0.37269864082336424,
      "mean_token_accuracy": 0.8786865592002868,
      "num_tokens": 35045028.0,
      "step": 630
    },
    {
      "entropy": 0.3767994062975049,
      "epoch": 0.9719058466211086,
      "grad_norm": 0.890625,
      "learning_rate": 2.516439250177749e-08,
      "loss": 0.3758098125457764,
      "mean_token_accuracy": 0.8778494797647,
      "num_tokens": 35600399.0,
      "step": 640
    },
    {
      "entropy": 0.3786265593022108,
      "epoch": 0.9870918754745635,
      "grad_norm": 0.6875,
      "learning_rate": 6.295060904623618e-09,
      "loss": 0.378217077255249,
      "mean_token_accuracy": 0.8778206452727317,
      "num_tokens": 36162293.0,
      "step": 650
    }
  ],
  "logging_steps": 10,
  "max_steps": 659,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.858614662906511e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}