{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.3009761388286334,
  "eval_steps": 500,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016268980477223426,
      "grad_norm": 5.2547173500061035,
      "learning_rate": 1.1062285714285713e-05,
      "loss": 1.7262,
      "mean_token_accuracy": 0.7495950579643249,
      "num_tokens": 630493.0,
      "step": 10
    },
    {
      "epoch": 0.03253796095444685,
      "grad_norm": 1.778620958328247,
      "learning_rate": 2.3353714285714284e-05,
      "loss": 0.4544,
      "mean_token_accuracy": 0.8820267458756764,
      "num_tokens": 1260394.0,
      "step": 20
    },
    {
      "epoch": 0.04880694143167028,
      "grad_norm": 1.519212007522583,
      "learning_rate": 3.564514285714286e-05,
      "loss": 0.1954,
      "mean_token_accuracy": 0.9297164340813955,
      "num_tokens": 1882452.0,
      "step": 30
    },
    {
      "epoch": 0.0650759219088937,
      "grad_norm": 1.317597508430481,
      "learning_rate": 4.301910802723618e-05,
      "loss": 0.1078,
      "mean_token_accuracy": 0.9594547629356385,
      "num_tokens": 2516165.0,
      "step": 40
    },
    {
      "epoch": 0.08134490238611713,
      "grad_norm": 0.9972535371780396,
      "learning_rate": 4.300907446637134e-05,
      "loss": 0.0749,
      "mean_token_accuracy": 0.9725557684898376,
      "num_tokens": 3140818.0,
      "step": 50
    },
    {
      "epoch": 0.09761388286334056,
      "grad_norm": 0.4797419607639313,
      "learning_rate": 4.298789933596705e-05,
      "loss": 0.0522,
      "mean_token_accuracy": 0.9811804493268331,
      "num_tokens": 3774733.0,
      "step": 60
    },
    {
      "epoch": 0.113882863340564,
      "grad_norm": 0.7913459539413452,
      "learning_rate": 4.295559727008953e-05,
      "loss": 0.0361,
      "mean_token_accuracy": 0.9873626848061879,
      "num_tokens": 4397570.0,
      "step": 70
    },
    {
      "epoch": 0.1301518438177874,
      "grad_norm": 1.19544517993927,
      "learning_rate": 4.291219059259517e-05,
      "loss": 0.0274,
      "mean_token_accuracy": 0.9901207347710927,
      "num_tokens": 5011799.0,
      "step": 80
    },
    {
      "epoch": 0.14642082429501085,
      "grad_norm": 0.9892869591712952,
      "learning_rate": 4.2857709301702605e-05,
      "loss": 0.0325,
      "mean_token_accuracy": 0.9884837865829468,
      "num_tokens": 5620525.0,
      "step": 90
    },
    {
      "epoch": 0.16268980477223427,
      "grad_norm": 0.4109046161174774,
      "learning_rate": 4.279219104926108e-05,
      "loss": 0.0285,
      "mean_token_accuracy": 0.9890055457750956,
      "num_tokens": 6232839.0,
      "step": 100
    },
    {
      "epoch": 0.1789587852494577,
      "grad_norm": 0.22267405688762665,
      "learning_rate": 4.2715681114729315e-05,
      "loss": 0.0216,
      "mean_token_accuracy": 0.99150417248408,
      "num_tokens": 6868903.0,
      "step": 110
    },
    {
      "epoch": 0.19522776572668113,
      "grad_norm": 0.7107319235801697,
      "learning_rate": 4.262823237388305e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9916373431682587,
      "num_tokens": 7484698.0,
      "step": 120
    },
    {
      "epoch": 0.21149674620390455,
      "grad_norm": 0.17150786519050598,
      "learning_rate": 4.2529905262272735e-05,
      "loss": 0.024,
      "mean_token_accuracy": 0.9917707979679108,
      "num_tokens": 8111052.0,
      "step": 130
    },
    {
      "epoch": 0.227765726681128,
      "grad_norm": 0.13397717475891113,
      "learning_rate": 4.242076773345677e-05,
      "loss": 0.0244,
      "mean_token_accuracy": 0.9901391108830769,
      "num_tokens": 8739411.0,
      "step": 140
    },
    {
      "epoch": 0.2440347071583514,
      "grad_norm": 0.5640156269073486,
      "learning_rate": 4.2300895212038937e-05,
      "loss": 0.0191,
      "mean_token_accuracy": 0.9922256747881572,
      "num_tokens": 9362736.0,
      "step": 150
    },
    {
      "epoch": 0.2603036876355748,
      "grad_norm": 0.23605239391326904,
      "learning_rate": 4.217037054154274e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9922493000825247,
      "num_tokens": 9987338.0,
      "step": 160
    },
    {
      "epoch": 0.2765726681127983,
      "grad_norm": 0.13247837126255035,
      "learning_rate": 4.2029283927158504e-05,
      "loss": 0.0236,
      "mean_token_accuracy": 0.9909312228361765,
      "num_tokens": 10597546.0,
      "step": 170
    },
    {
      "epoch": 0.2928416485900217,
      "grad_norm": 0.16422192752361298,
      "learning_rate": 4.187773287340283e-05,
      "loss": 0.018,
      "mean_token_accuracy": 0.9934214214483897,
      "num_tokens": 11220635.0,
      "step": 180
    },
    {
      "epoch": 0.3091106290672451,
      "grad_norm": 0.2876814901828766,
      "learning_rate": 4.171582211673354e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9923185408115387,
      "num_tokens": 11844329.0,
      "step": 190
    },
    {
      "epoch": 0.32537960954446854,
      "grad_norm": 0.25289878249168396,
      "learning_rate": 4.154366355316665e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.9930294732252757,
      "num_tokens": 12455864.0,
      "step": 200
    },
    {
      "epoch": 0.34164859002169196,
      "grad_norm": 0.25638192892074585,
      "learning_rate": 4.136137616094537e-05,
      "loss": 0.0195,
      "mean_token_accuracy": 0.9926043192545573,
      "num_tokens": 13090837.0,
      "step": 210
    },
    {
      "epoch": 0.3579175704989154,
      "grad_norm": 0.10495291650295258,
      "learning_rate": 4.1169085918314616e-05,
      "loss": 0.0198,
      "mean_token_accuracy": 0.9918980141480763,
      "num_tokens": 13714626.0,
      "step": 220
    },
    {
      "epoch": 0.37418655097613884,
      "grad_norm": 0.09805836528539658,
      "learning_rate": 4.096692571645784e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9913817584514618,
      "num_tokens": 14336385.0,
      "step": 230
    },
    {
      "epoch": 0.39045553145336226,
      "grad_norm": 0.27117225527763367,
      "learning_rate": 4.075503526765637e-05,
      "loss": 0.021,
      "mean_token_accuracy": 0.9919139047463735,
      "num_tokens": 14960133.0,
      "step": 240
    },
    {
      "epoch": 0.4067245119305857,
      "grad_norm": 0.10686483234167099,
      "learning_rate": 4.053356100873465e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9918312350908915,
      "num_tokens": 15576028.0,
      "step": 250
    },
    {
      "epoch": 0.4229934924078091,
      "grad_norm": 0.4648778438568115,
      "learning_rate": 4.030265599985825e-05,
      "loss": 0.0202,
      "mean_token_accuracy": 0.9926771819591522,
      "num_tokens": 16195394.0,
      "step": 260
    },
    {
      "epoch": 0.43926247288503256,
      "grad_norm": 0.180766761302948,
      "learning_rate": 4.0062479818754434e-05,
      "loss": 0.0192,
      "mean_token_accuracy": 0.9923730830351511,
      "num_tokens": 16834203.0,
      "step": 270
    },
    {
      "epoch": 0.455531453362256,
      "grad_norm": 0.06381647288799286,
      "learning_rate": 3.981319845042853e-05,
      "loss": 0.0196,
      "mean_token_accuracy": 0.9922918160756429,
      "num_tokens": 17456060.0,
      "step": 280
    },
    {
      "epoch": 0.4718004338394794,
      "grad_norm": 0.17192164063453674,
      "learning_rate": 3.955498417245217e-05,
      "loss": 0.0233,
      "mean_token_accuracy": 0.9919982612133026,
      "num_tokens": 18084365.0,
      "step": 290
    },
    {
      "epoch": 0.4880694143167028,
      "grad_norm": 0.05631360784173012,
      "learning_rate": 3.928801543590283e-05,
      "loss": 0.0172,
      "mean_token_accuracy": 0.9939788281917572,
      "num_tokens": 18725183.0,
      "step": 300
    },
    {
      "epoch": 0.5043383947939263,
      "grad_norm": 0.2000889927148819,
      "learning_rate": 3.9012476742036844e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.9914020558198293,
      "num_tokens": 19348901.0,
      "step": 310
    },
    {
      "epoch": 0.5206073752711496,
      "grad_norm": 0.07515931874513626,
      "learning_rate": 3.872855851478117e-05,
      "loss": 0.0171,
      "mean_token_accuracy": 0.9934311370054881,
      "num_tokens": 19978469.0,
      "step": 320
    },
    {
      "epoch": 0.5368763557483731,
      "grad_norm": 0.09410631656646729,
      "learning_rate": 3.8436456969131984e-05,
      "loss": 0.0227,
      "mean_token_accuracy": 0.9908039093017578,
      "num_tokens": 20599645.0,
      "step": 330
    },
    {
      "epoch": 0.5531453362255966,
      "grad_norm": 0.039398010820150375,
      "learning_rate": 3.813637397555116e-05,
      "loss": 0.0175,
      "mean_token_accuracy": 0.9934881230195364,
      "num_tokens": 21222229.0,
      "step": 340
    },
    {
      "epoch": 0.56941431670282,
      "grad_norm": 0.07333279401063919,
      "learning_rate": 3.782851692045417e-05,
      "loss": 0.0229,
      "mean_token_accuracy": 0.9910738269488016,
      "num_tokens": 21845972.0,
      "step": 350
    },
    {
      "epoch": 0.5856832971800434,
      "grad_norm": 0.08670751005411148,
      "learning_rate": 3.751309856288604e-05,
      "loss": 0.0167,
      "mean_token_accuracy": 0.9931929151217143,
      "num_tokens": 22473822.0,
      "step": 360
    },
    {
      "epoch": 0.6019522776572668,
      "grad_norm": 0.07762621343135834,
      "learning_rate": 3.719033688748419e-05,
      "loss": 0.0155,
      "mean_token_accuracy": 0.9943595806757609,
      "num_tokens": 23108873.0,
      "step": 370
    },
    {
      "epoch": 0.6182212581344902,
      "grad_norm": 0.09424326568841934,
      "learning_rate": 3.6860454953829976e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9918720722198486,
      "num_tokens": 23735218.0,
      "step": 380
    },
    {
      "epoch": 0.6344902386117137,
      "grad_norm": 0.07514937967061996,
      "learning_rate": 3.652368074229288e-05,
      "loss": 0.0169,
      "mean_token_accuracy": 0.9938062051932017,
      "num_tokens": 24358846.0,
      "step": 390
    },
    {
      "epoch": 0.6507592190889371,
      "grad_norm": 0.06997387111186981,
      "learning_rate": 3.618024699647403e-05,
      "loss": 0.0194,
      "mean_token_accuracy": 0.9927034497261047,
      "num_tokens": 24970185.0,
      "step": 400
    },
    {
      "epoch": 0.6670281995661606,
      "grad_norm": 0.04538058117032051,
      "learning_rate": 3.5830391062357777e-05,
      "loss": 0.0171,
      "mean_token_accuracy": 0.9933724681536357,
      "num_tokens": 25610071.0,
      "step": 410
    },
    {
      "epoch": 0.6832971800433839,
      "grad_norm": 0.07518076151609421,
      "learning_rate": 3.547435472428268e-05,
      "loss": 0.0183,
      "mean_token_accuracy": 0.9926396747430165,
      "num_tokens": 26225327.0,
      "step": 420
    },
    {
      "epoch": 0.6995661605206074,
      "grad_norm": 0.04776303842663765,
      "learning_rate": 3.5112384037845075e-05,
      "loss": 0.0177,
      "mean_token_accuracy": 0.9932779908180237,
      "num_tokens": 26845894.0,
      "step": 430
    },
    {
      "epoch": 0.7158351409978309,
      "grad_norm": 0.0617113932967186,
      "learning_rate": 3.4744729159850823e-05,
      "loss": 0.0165,
      "mean_token_accuracy": 0.9937874178091685,
      "num_tokens": 27456901.0,
      "step": 440
    },
    {
      "epoch": 0.7321041214750542,
      "grad_norm": 0.03345214203000069,
      "learning_rate": 3.437164417543277e-05,
      "loss": 0.0164,
      "mean_token_accuracy": 0.9934410572052002,
      "num_tokens": 28075053.0,
      "step": 450
    },
    {
      "epoch": 0.7483731019522777,
      "grad_norm": 0.06391480565071106,
      "learning_rate": 3.3993386922453256e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9922087252140045,
      "num_tokens": 28698113.0,
      "step": 460
    },
    {
      "epoch": 0.764642082429501,
      "grad_norm": 0.08372132480144501,
      "learning_rate": 3.36102188133132e-05,
      "loss": 0.0176,
      "mean_token_accuracy": 0.9931006868680318,
      "num_tokens": 29331343.0,
      "step": 470
    },
    {
      "epoch": 0.7809110629067245,
      "grad_norm": 0.05750415846705437,
      "learning_rate": 3.3222404654290796e-05,
      "loss": 0.0167,
      "mean_token_accuracy": 0.9928392628828685,
      "num_tokens": 29955389.0,
      "step": 480
    },
    {
      "epoch": 0.797180043383948,
      "grad_norm": 0.0684645026922226,
      "learning_rate": 3.283021246253477e-05,
      "loss": 0.0191,
      "mean_token_accuracy": 0.9932359317938487,
      "num_tokens": 30582110.0,
      "step": 490
    },
    {
      "epoch": 0.8134490238611713,
      "grad_norm": 0.05831017345190048,
      "learning_rate": 3.24339132808385e-05,
      "loss": 0.0211,
      "mean_token_accuracy": 0.9914661645889282,
      "num_tokens": 31199929.0,
      "step": 500
    },
    {
      "epoch": 0.8297180043383948,
      "grad_norm": 0.05367962643504143,
      "learning_rate": 3.203378099032326e-05,
      "loss": 0.0168,
      "mean_token_accuracy": 0.9932946304480235,
      "num_tokens": 31834372.0,
      "step": 510
    },
    {
      "epoch": 0.8459869848156182,
      "grad_norm": 0.07528548687696457,
      "learning_rate": 3.163009212115977e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9925516506036123,
      "num_tokens": 32458856.0,
      "step": 520
    },
    {
      "epoch": 0.8622559652928417,
      "grad_norm": 0.05971767008304596,
      "learning_rate": 3.122312566145912e-05,
      "loss": 0.0187,
      "mean_token_accuracy": 0.9925659577051799,
      "num_tokens": 33089190.0,
      "step": 530
    },
    {
      "epoch": 0.8785249457700651,
      "grad_norm": 0.072539322078228,
      "learning_rate": 3.081316286446494e-05,
      "loss": 0.0183,
      "mean_token_accuracy": 0.9925340354442597,
      "num_tokens": 33694708.0,
      "step": 540
    },
    {
      "epoch": 0.8947939262472885,
      "grad_norm": 0.08411821722984314,
      "learning_rate": 3.0400487054180153e-05,
      "loss": 0.0212,
      "mean_token_accuracy": 0.9916836241881053,
      "num_tokens": 34307427.0,
      "step": 550
    },
    {
      "epoch": 0.911062906724512,
      "grad_norm": 0.150554358959198,
      "learning_rate": 2.9985383429562652e-05,
      "loss": 0.02,
      "mean_token_accuracy": 0.9929757992426554,
      "num_tokens": 34927922.0,
      "step": 560
    },
    {
      "epoch": 0.9273318872017353,
      "grad_norm": 0.02936033345758915,
      "learning_rate": 2.956813886742522e-05,
      "loss": 0.0172,
      "mean_token_accuracy": 0.9933930238087972,
      "num_tokens": 35557289.0,
      "step": 570
    },
    {
      "epoch": 0.9436008676789588,
      "grad_norm": 0.056036002933979034,
      "learning_rate": 2.914904172417582e-05,
      "loss": 0.0151,
      "mean_token_accuracy": 0.9941118697325388,
      "num_tokens": 36189524.0,
      "step": 580
    },
    {
      "epoch": 0.9598698481561823,
      "grad_norm": 0.08125586807727814,
      "learning_rate": 2.8728381636535426e-05,
      "loss": 0.0175,
      "mean_token_accuracy": 0.9934323211510976,
      "num_tokens": 36823151.0,
      "step": 590
    },
    {
      "epoch": 0.9761388286334056,
      "grad_norm": 0.044415201991796494,
      "learning_rate": 2.8306449321370966e-05,
      "loss": 0.0156,
      "mean_token_accuracy": 0.9935656964778901,
      "num_tokens": 37454804.0,
      "step": 600
    },
    {
      "epoch": 0.9924078091106291,
      "grad_norm": 0.05092897638678551,
      "learning_rate": 2.788353637478183e-05,
      "loss": 0.0183,
      "mean_token_accuracy": 0.9926630795001984,
      "num_tokens": 38066684.0,
      "step": 610
    },
    {
      "epoch": 1.0081344902386118,
      "grad_norm": 0.06339436024427414,
      "learning_rate": 2.7459935070578716e-05,
      "loss": 0.0181,
      "mean_token_accuracy": 0.9929262954613258,
      "num_tokens": 38668867.0,
      "step": 620
    },
    {
      "epoch": 1.0244034707158352,
      "grad_norm": 0.061983563005924225,
      "learning_rate": 2.7035938158294073e-05,
      "loss": 0.0178,
      "mean_token_accuracy": 0.9929542601108551,
      "num_tokens": 39295365.0,
      "step": 630
    },
    {
      "epoch": 1.0406724511930585,
      "grad_norm": 0.05101089924573898,
      "learning_rate": 2.6611838660863893e-05,
      "loss": 0.0186,
      "mean_token_accuracy": 0.9923582990964254,
      "num_tokens": 39906855.0,
      "step": 640
    },
    {
      "epoch": 1.056941431670282,
      "grad_norm": 0.05050944909453392,
      "learning_rate": 2.6187929672120427e-05,
      "loss": 0.0162,
      "mean_token_accuracy": 0.9933266679445902,
      "num_tokens": 40530351.0,
      "step": 650
    },
    {
      "epoch": 1.0732104121475055,
      "grad_norm": 0.038513652980327606,
      "learning_rate": 2.5764504154235977e-05,
      "loss": 0.0185,
      "mean_token_accuracy": 0.9930866738160451,
      "num_tokens": 41155436.0,
      "step": 660
    },
    {
      "epoch": 1.0894793926247288,
      "grad_norm": 0.05838533863425255,
      "learning_rate": 2.5341854735257622e-05,
      "loss": 0.0165,
      "mean_token_accuracy": 0.99360005458196,
      "num_tokens": 41782248.0,
      "step": 670
    },
    {
      "epoch": 1.1057483731019522,
      "grad_norm": 0.08960133790969849,
      "learning_rate": 2.4920273506872954e-05,
      "loss": 0.0147,
      "mean_token_accuracy": 0.9939835667610168,
      "num_tokens": 42419401.0,
      "step": 680
    },
    {
      "epoch": 1.1220173535791758,
      "grad_norm": 0.06357277184724808,
      "learning_rate": 2.450005182254639e-05,
      "loss": 0.0171,
      "mean_token_accuracy": 0.993314818541209,
      "num_tokens": 43039973.0,
      "step": 690
    },
    {
      "epoch": 1.1382863340563991,
      "grad_norm": 0.051391322165727615,
      "learning_rate": 2.408148009616571e-05,
      "loss": 0.0197,
      "mean_token_accuracy": 0.9920130769411722,
      "num_tokens": 43661330.0,
      "step": 700
    },
    {
      "epoch": 1.1545553145336225,
      "grad_norm": 0.22426648437976837,
      "learning_rate": 2.366484760133793e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9920180996259054,
      "num_tokens": 44283330.0,
      "step": 710
    },
    {
      "epoch": 1.170824295010846,
      "grad_norm": 0.06203395873308182,
      "learning_rate": 2.325044227147322e-05,
      "loss": 0.0156,
      "mean_token_accuracy": 0.993381271759669,
      "num_tokens": 44904744.0,
      "step": 720
    },
    {
      "epoch": 1.1870932754880694,
      "grad_norm": 0.07929422706365585,
      "learning_rate": 2.2838550500795045e-05,
      "loss": 0.0193,
      "mean_token_accuracy": 0.9923672596613566,
      "num_tokens": 45522192.0,
      "step": 730
    },
    {
      "epoch": 1.2033622559652928,
      "grad_norm": 0.02930581010878086,
      "learning_rate": 2.2429456946413984e-05,
      "loss": 0.0189,
      "mean_token_accuracy": 0.992294559876124,
      "num_tokens": 46149764.0,
      "step": 740
    },
    {
      "epoch": 1.2196312364425164,
      "grad_norm": 0.08023188263177872,
      "learning_rate": 2.2023444331602134e-05,
      "loss": 0.0208,
      "mean_token_accuracy": 0.9917561610539755,
      "num_tokens": 46753101.0,
      "step": 750
    },
    {
      "epoch": 1.2359002169197397,
      "grad_norm": 0.06463509052991867,
      "learning_rate": 2.1620793250403904e-05,
      "loss": 0.0174,
      "mean_token_accuracy": 0.992903612057368,
      "num_tokens": 47370578.0,
      "step": 760
    },
    {
      "epoch": 1.252169197396963,
      "grad_norm": 0.07925821095705032,
      "learning_rate": 2.1221781973718358e-05,
      "loss": 0.0181,
      "mean_token_accuracy": 0.9931439737478892,
      "num_tokens": 47994002.0,
      "step": 770
    },
    {
      "epoch": 1.2684381778741867,
      "grad_norm": 0.05930013209581375,
      "learning_rate": 2.0826686256987106e-05,
      "loss": 0.0199,
      "mean_token_accuracy": 0.991779484351476,
      "num_tokens": 48608280.0,
      "step": 780
    },
    {
      "epoch": 1.28470715835141,
      "grad_norm": 0.05656487122178078,
      "learning_rate": 2.0435779149620534e-05,
      "loss": 0.0157,
      "mean_token_accuracy": 0.9936818579832712,
      "num_tokens": 49243847.0,
      "step": 790
    },
    {
      "epoch": 1.3009761388286334,
      "grad_norm": 0.05386212468147278,
      "learning_rate": 2.0049330806294254e-05,
      "loss": 0.0178,
      "mean_token_accuracy": 0.9924297213554383,
      "num_tokens": 49855791.0,
      "step": 800
    }
  ],
  "logging_steps": 10,
  "max_steps": 1230,
  "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": 2.7877236939864146e+18,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}