{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 168,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "aux_loss": 8.045117378234863,
      "entropy": 0.05226727016270161,
      "epoch": 0.024096385542168676,
      "grad_norm": 0.5021907687187195,
      "learning_rate": 0.0,
      "loss": 0.09741637855768204,
      "mean_token_accuracy": 0.9888820052146912,
      "num_tokens": 154362.0,
      "step": 1
    },
    {
      "aux_loss": 8.047614574432373,
      "entropy": 0.05835266411304474,
      "epoch": 0.04819277108433735,
      "grad_norm": 0.551396369934082,
      "learning_rate": 1.4500000000000001e-07,
      "loss": 0.09991507232189178,
      "mean_token_accuracy": 0.9883801937103271,
      "num_tokens": 310702.0,
      "step": 2
    },
    {
      "aux_loss": 8.049727439880371,
      "entropy": 0.052583809942007065,
      "epoch": 0.07228915662650602,
      "grad_norm": 0.4750031530857086,
      "learning_rate": 2.9000000000000003e-07,
      "loss": 0.09806497395038605,
      "mean_token_accuracy": 0.9886206984519958,
      "num_tokens": 493904.0,
      "step": 3
    },
    {
      "aux_loss": 8.047826766967773,
      "entropy": 0.05231792479753494,
      "epoch": 0.0963855421686747,
      "grad_norm": 0.4863564074039459,
      "learning_rate": 4.35e-07,
      "loss": 0.09776315838098526,
      "mean_token_accuracy": 0.9888334274291992,
      "num_tokens": 663136.0,
      "step": 4
    },
    {
      "aux_loss": 8.047198295593262,
      "entropy": 0.04943128302693367,
      "epoch": 0.12048192771084337,
      "grad_norm": 0.5010114908218384,
      "learning_rate": 5.800000000000001e-07,
      "loss": 0.09543517231941223,
      "mean_token_accuracy": 0.9894405901432037,
      "num_tokens": 812532.0,
      "step": 5
    },
    {
      "aux_loss": 8.049106121063232,
      "entropy": 0.063042301684618,
      "epoch": 0.14457831325301204,
      "grad_norm": 0.5165722966194153,
      "learning_rate": 7.25e-07,
      "loss": 0.10346449911594391,
      "mean_token_accuracy": 0.987544983625412,
      "num_tokens": 973071.0,
      "step": 6
    },
    {
      "aux_loss": 8.047893047332764,
      "entropy": 0.0672433041036129,
      "epoch": 0.1686746987951807,
      "grad_norm": 0.5849524736404419,
      "learning_rate": 8.7e-07,
      "loss": 0.10626787692308426,
      "mean_token_accuracy": 0.9862557649612427,
      "num_tokens": 1110326.0,
      "step": 7
    },
    {
      "aux_loss": 8.04862642288208,
      "entropy": 0.06088166497647762,
      "epoch": 0.1927710843373494,
      "grad_norm": 0.48005586862564087,
      "learning_rate": 1.015e-06,
      "loss": 0.1021210104227066,
      "mean_token_accuracy": 0.9877255260944366,
      "num_tokens": 1259440.0,
      "step": 8
    },
    {
      "aux_loss": 8.047185897827148,
      "entropy": 0.049826065078377724,
      "epoch": 0.21686746987951808,
      "grad_norm": 0.43366947770118713,
      "learning_rate": 1.1600000000000001e-06,
      "loss": 0.09475715458393097,
      "mean_token_accuracy": 0.9898165464401245,
      "num_tokens": 1417756.0,
      "step": 9
    },
    {
      "aux_loss": 8.04772663116455,
      "entropy": 0.054957231506705284,
      "epoch": 0.24096385542168675,
      "grad_norm": 0.4529463052749634,
      "learning_rate": 1.3050000000000002e-06,
      "loss": 0.09896557033061981,
      "mean_token_accuracy": 0.9889253675937653,
      "num_tokens": 1579862.0,
      "step": 10
    },
    {
      "aux_loss": 8.047585487365723,
      "entropy": 0.05796742998063564,
      "epoch": 0.26506024096385544,
      "grad_norm": 0.45284125208854675,
      "learning_rate": 1.45e-06,
      "loss": 0.10001235455274582,
      "mean_token_accuracy": 0.9876790642738342,
      "num_tokens": 1731184.0,
      "step": 11
    },
    {
      "aux_loss": 8.04898738861084,
      "entropy": 0.05215233750641346,
      "epoch": 0.2891566265060241,
      "grad_norm": 0.4420299828052521,
      "learning_rate": 1.4499995904705798e-06,
      "loss": 0.09724222123622894,
      "mean_token_accuracy": 0.9884337782859802,
      "num_tokens": 1893932.0,
      "step": 12
    },
    {
      "aux_loss": 8.04798412322998,
      "entropy": 0.05265074409544468,
      "epoch": 0.3132530120481928,
      "grad_norm": 0.40204742550849915,
      "learning_rate": 1.4499983618832442e-06,
      "loss": 0.0970243290066719,
      "mean_token_accuracy": 0.989496260881424,
      "num_tokens": 2053005.0,
      "step": 13
    },
    {
      "aux_loss": 8.05030107498169,
      "entropy": 0.049178898334503174,
      "epoch": 0.3373493975903614,
      "grad_norm": 0.4108814597129822,
      "learning_rate": 1.4499963142407692e-06,
      "loss": 0.09499870240688324,
      "mean_token_accuracy": 0.9897345900535583,
      "num_tokens": 2217039.0,
      "step": 14
    },
    {
      "aux_loss": 8.046091079711914,
      "entropy": 0.0504351407289505,
      "epoch": 0.3614457831325301,
      "grad_norm": 0.4030505418777466,
      "learning_rate": 1.4499934475477814e-06,
      "loss": 0.09597203135490417,
      "mean_token_accuracy": 0.9898054599761963,
      "num_tokens": 2386769.0,
      "step": 15
    },
    {
      "aux_loss": 8.047934532165527,
      "entropy": 0.04648676887154579,
      "epoch": 0.3855421686746988,
      "grad_norm": 0.4123123586177826,
      "learning_rate": 1.4499897618107582e-06,
      "loss": 0.09345895051956177,
      "mean_token_accuracy": 0.989924967288971,
      "num_tokens": 2551093.0,
      "step": 16
    },
    {
      "aux_loss": 8.04793405532837,
      "entropy": 0.04946734011173248,
      "epoch": 0.40963855421686746,
      "grad_norm": 0.42739129066467285,
      "learning_rate": 1.4499852570380272e-06,
      "loss": 0.09633760154247284,
      "mean_token_accuracy": 0.9889810979366302,
      "num_tokens": 2704860.0,
      "step": 17
    },
    {
      "aux_loss": 8.048253536224365,
      "entropy": 0.0546275470405817,
      "epoch": 0.43373493975903615,
      "grad_norm": 0.44881728291511536,
      "learning_rate": 1.449979933239767e-06,
      "loss": 0.09820500761270523,
      "mean_token_accuracy": 0.9885384142398834,
      "num_tokens": 2850860.0,
      "step": 18
    },
    {
      "aux_loss": 8.050773620605469,
      "entropy": 0.05041976273059845,
      "epoch": 0.4578313253012048,
      "grad_norm": 0.46621593832969666,
      "learning_rate": 1.4499737904280063e-06,
      "loss": 0.09671641141176224,
      "mean_token_accuracy": 0.9887768924236298,
      "num_tokens": 3039323.0,
      "step": 19
    },
    {
      "aux_loss": 8.050614356994629,
      "entropy": 0.050554944202303886,
      "epoch": 0.4819277108433735,
      "grad_norm": 0.47147253155708313,
      "learning_rate": 1.449966828616625e-06,
      "loss": 0.09628961980342865,
      "mean_token_accuracy": 0.9888468682765961,
      "num_tokens": 3195488.0,
      "step": 20
    },
    {
      "aux_loss": 8.050609111785889,
      "entropy": 0.04489951767027378,
      "epoch": 0.5060240963855421,
      "grad_norm": 0.4890425503253937,
      "learning_rate": 1.4499590478213525e-06,
      "loss": 0.09394823014736176,
      "mean_token_accuracy": 0.9896246194839478,
      "num_tokens": 3369627.0,
      "step": 21
    },
    {
      "aux_loss": 8.047994613647461,
      "entropy": 0.0530005544424057,
      "epoch": 0.5301204819277109,
      "grad_norm": 0.5181686878204346,
      "learning_rate": 1.4499504480597702e-06,
      "loss": 0.09846185147762299,
      "mean_token_accuracy": 0.9879846572875977,
      "num_tokens": 3520643.0,
      "step": 22
    },
    {
      "aux_loss": 8.049047946929932,
      "entropy": 0.039870962500572205,
      "epoch": 0.5542168674698795,
      "grad_norm": 0.41603270173072815,
      "learning_rate": 1.4499410293513081e-06,
      "loss": 0.09075028449296951,
      "mean_token_accuracy": 0.9908623099327087,
      "num_tokens": 3702510.0,
      "step": 23
    },
    {
      "aux_loss": 8.050177097320557,
      "entropy": 0.05053388141095638,
      "epoch": 0.5783132530120482,
      "grad_norm": 0.4234795570373535,
      "learning_rate": 1.449930791717248e-06,
      "loss": 0.09689382463693619,
      "mean_token_accuracy": 0.9886369109153748,
      "num_tokens": 3867610.0,
      "step": 24
    },
    {
      "aux_loss": 8.048462867736816,
      "entropy": 0.049235520884394646,
      "epoch": 0.6024096385542169,
      "grad_norm": 0.4375890791416168,
      "learning_rate": 1.4499197351807219e-06,
      "loss": 0.09504671394824982,
      "mean_token_accuracy": 0.9891063868999481,
      "num_tokens": 4021186.0,
      "step": 25
    },
    {
      "aux_loss": 8.050186157226562,
      "entropy": 0.048021236434578896,
      "epoch": 0.6265060240963856,
      "grad_norm": 0.42389488220214844,
      "learning_rate": 1.449907859766711e-06,
      "loss": 0.09462042152881622,
      "mean_token_accuracy": 0.9893973469734192,
      "num_tokens": 4183102.0,
      "step": 26
    },
    {
      "aux_loss": 8.04945182800293,
      "entropy": 0.042689668014645576,
      "epoch": 0.6506024096385542,
      "grad_norm": 0.40038207173347473,
      "learning_rate": 1.4498951655020479e-06,
      "loss": 0.09214234352111816,
      "mean_token_accuracy": 0.9904394745826721,
      "num_tokens": 4348628.0,
      "step": 27
    },
    {
      "aux_loss": 8.050597667694092,
      "entropy": 0.05143065005540848,
      "epoch": 0.6746987951807228,
      "grad_norm": 0.4422961473464966,
      "learning_rate": 1.4498816524154148e-06,
      "loss": 0.09722501039505005,
      "mean_token_accuracy": 0.9888282418251038,
      "num_tokens": 4511217.0,
      "step": 28
    },
    {
      "aux_loss": 8.048704147338867,
      "entropy": 0.044764744117856026,
      "epoch": 0.6987951807228916,
      "grad_norm": 0.4279940128326416,
      "learning_rate": 1.4498673205373443e-06,
      "loss": 0.09300120174884796,
      "mean_token_accuracy": 0.9899511337280273,
      "num_tokens": 4684266.0,
      "step": 29
    },
    {
      "aux_loss": 8.047113418579102,
      "entropy": 0.04995520040392876,
      "epoch": 0.7228915662650602,
      "grad_norm": 0.45497408509254456,
      "learning_rate": 1.4498521699002187e-06,
      "loss": 0.0966445729136467,
      "mean_token_accuracy": 0.9883378744125366,
      "num_tokens": 4858009.0,
      "step": 30
    },
    {
      "aux_loss": 8.050236701965332,
      "entropy": 0.04773502238094807,
      "epoch": 0.7469879518072289,
      "grad_norm": 0.40059739351272583,
      "learning_rate": 1.4498362005382707e-06,
      "loss": 0.09425598382949829,
      "mean_token_accuracy": 0.9895853698253632,
      "num_tokens": 5041560.0,
      "step": 31
    },
    {
      "aux_loss": 8.0483078956604,
      "entropy": 0.04463958367705345,
      "epoch": 0.7710843373493976,
      "grad_norm": 0.421403706073761,
      "learning_rate": 1.4498194124875822e-06,
      "loss": 0.0921817421913147,
      "mean_token_accuracy": 0.9903146624565125,
      "num_tokens": 5207925.0,
      "step": 32
    },
    {
      "aux_loss": 8.051517486572266,
      "entropy": 0.0455746091902256,
      "epoch": 0.7951807228915663,
      "grad_norm": 0.39467114210128784,
      "learning_rate": 1.449801805786086e-06,
      "loss": 0.09311801195144653,
      "mean_token_accuracy": 0.9897293448448181,
      "num_tokens": 5410900.0,
      "step": 33
    },
    {
      "aux_loss": 8.047285079956055,
      "entropy": 0.04475598596036434,
      "epoch": 0.8192771084337349,
      "grad_norm": 0.3885505795478821,
      "learning_rate": 1.4497833804735631e-06,
      "loss": 0.09283807128667831,
      "mean_token_accuracy": 0.9904457926750183,
      "num_tokens": 5579379.0,
      "step": 34
    },
    {
      "aux_loss": 8.049206733703613,
      "entropy": 0.045768845826387405,
      "epoch": 0.8433734939759037,
      "grad_norm": 0.38822969794273376,
      "learning_rate": 1.4497641365916457e-06,
      "loss": 0.0930660218000412,
      "mean_token_accuracy": 0.9898505806922913,
      "num_tokens": 5756181.0,
      "step": 35
    },
    {
      "aux_loss": 8.049548149108887,
      "entropy": 0.0442772526293993,
      "epoch": 0.8674698795180723,
      "grad_norm": 0.42765918374061584,
      "learning_rate": 1.4497440741838147e-06,
      "loss": 0.09213966131210327,
      "mean_token_accuracy": 0.9906859695911407,
      "num_tokens": 5913888.0,
      "step": 36
    },
    {
      "aux_loss": 8.047310829162598,
      "entropy": 0.05137152969837189,
      "epoch": 0.891566265060241,
      "grad_norm": 0.46749433875083923,
      "learning_rate": 1.4497231932954002e-06,
      "loss": 0.09635019302368164,
      "mean_token_accuracy": 0.9894134104251862,
      "num_tokens": 6073822.0,
      "step": 37
    },
    {
      "aux_loss": 8.04575777053833,
      "entropy": 0.04952680319547653,
      "epoch": 0.9156626506024096,
      "grad_norm": 0.40102288126945496,
      "learning_rate": 1.4497014939735825e-06,
      "loss": 0.09520888328552246,
      "mean_token_accuracy": 0.9900149703025818,
      "num_tokens": 6228998.0,
      "step": 38
    },
    {
      "aux_loss": 8.05048656463623,
      "entropy": 0.05141362361609936,
      "epoch": 0.9397590361445783,
      "grad_norm": 0.45891159772872925,
      "learning_rate": 1.4496789762673903e-06,
      "loss": 0.09608915448188782,
      "mean_token_accuracy": 0.9895050525665283,
      "num_tokens": 6380479.0,
      "step": 39
    },
    {
      "aux_loss": 8.049390316009521,
      "entropy": 0.0542642492800951,
      "epoch": 0.963855421686747,
      "grad_norm": 0.4678654968738556,
      "learning_rate": 1.4496556402277018e-06,
      "loss": 0.0986521765589714,
      "mean_token_accuracy": 0.9882148206233978,
      "num_tokens": 6514347.0,
      "step": 40
    },
    {
      "aux_loss": 8.047286033630371,
      "entropy": 0.04129311256110668,
      "epoch": 0.9879518072289156,
      "grad_norm": 0.42529988288879395,
      "learning_rate": 1.4496314859072443e-06,
      "loss": 0.09138549864292145,
      "mean_token_accuracy": 0.9908298254013062,
      "num_tokens": 6684739.0,
      "step": 41
    },
    {
      "aux_loss": 8.050637245178223,
      "entropy": 0.04988313093781471,
      "epoch": 1.0,
      "grad_norm": 0.5628013610839844,
      "learning_rate": 1.4496065133605937e-06,
      "loss": 0.06163128837943077,
      "mean_token_accuracy": 0.9887231588363647,
      "num_tokens": 6741463.0,
      "step": 42
    },
    {
      "aux_loss": 8.047677040100098,
      "entropy": 0.04833286814391613,
      "epoch": 1.0240963855421688,
      "grad_norm": 0.43433839082717896,
      "learning_rate": 1.4495807226441745e-06,
      "loss": 0.09322302043437958,
      "mean_token_accuracy": 0.9907335042953491,
      "num_tokens": 6872162.0,
      "step": 43
    },
    {
      "aux_loss": 8.049845218658447,
      "entropy": 0.04814961925148964,
      "epoch": 1.0481927710843373,
      "grad_norm": 0.38683706521987915,
      "learning_rate": 1.4495541138162606e-06,
      "loss": 0.09274052828550339,
      "mean_token_accuracy": 0.9911822378635406,
      "num_tokens": 7044917.0,
      "step": 44
    },
    {
      "aux_loss": 8.045840740203857,
      "entropy": 0.04862277954816818,
      "epoch": 1.072289156626506,
      "grad_norm": 0.43781232833862305,
      "learning_rate": 1.4495266869369737e-06,
      "loss": 0.09285125136375427,
      "mean_token_accuracy": 0.991097629070282,
      "num_tokens": 7207551.0,
      "step": 45
    },
    {
      "aux_loss": 8.047110080718994,
      "entropy": 0.04816548526287079,
      "epoch": 1.0963855421686748,
      "grad_norm": 0.4117249846458435,
      "learning_rate": 1.4494984420682841e-06,
      "loss": 0.09330926835536957,
      "mean_token_accuracy": 0.990231990814209,
      "num_tokens": 7376391.0,
      "step": 46
    },
    {
      "aux_loss": 8.046239852905273,
      "entropy": 0.04640440456569195,
      "epoch": 1.1204819277108433,
      "grad_norm": 0.3918110728263855,
      "learning_rate": 1.4494693792740101e-06,
      "loss": 0.09266579151153564,
      "mean_token_accuracy": 0.9905861914157867,
      "num_tokens": 7544885.0,
      "step": 47
    },
    {
      "aux_loss": 8.045721054077148,
      "entropy": 0.04254587180912495,
      "epoch": 1.144578313253012,
      "grad_norm": 0.3635147213935852,
      "learning_rate": 1.4494394986198189e-06,
      "loss": 0.08964599668979645,
      "mean_token_accuracy": 0.9919890761375427,
      "num_tokens": 7718825.0,
      "step": 48
    },
    {
      "aux_loss": 8.04783582687378,
      "entropy": 0.0457068495452404,
      "epoch": 1.1686746987951806,
      "grad_norm": 0.37491077184677124,
      "learning_rate": 1.4494088001732242e-06,
      "loss": 0.0912659764289856,
      "mean_token_accuracy": 0.9914940893650055,
      "num_tokens": 7884797.0,
      "step": 49
    },
    {
      "aux_loss": 8.048118591308594,
      "entropy": 0.05809796415269375,
      "epoch": 1.1927710843373494,
      "grad_norm": 0.442947119474411,
      "learning_rate": 1.449377284003589e-06,
      "loss": 0.09830181300640106,
      "mean_token_accuracy": 0.9896205961704254,
      "num_tokens": 8037756.0,
      "step": 50
    },
    {
      "aux_loss": 8.045411109924316,
      "entropy": 0.04734855704009533,
      "epoch": 1.216867469879518,
      "grad_norm": 0.3778594732284546,
      "learning_rate": 1.449344950182123e-06,
      "loss": 0.09220214188098907,
      "mean_token_accuracy": 0.9912082552909851,
      "num_tokens": 8201900.0,
      "step": 51
    },
    {
      "aux_loss": 8.051240921020508,
      "entropy": 0.04616173356771469,
      "epoch": 1.2409638554216866,
      "grad_norm": 0.39906826615333557,
      "learning_rate": 1.4493117987818833e-06,
      "loss": 0.09254124015569687,
      "mean_token_accuracy": 0.9910232424736023,
      "num_tokens": 8368482.0,
      "step": 52
    },
    {
      "aux_loss": 8.048255443572998,
      "entropy": 0.04362104274332523,
      "epoch": 1.2650602409638554,
      "grad_norm": 0.4132522642612457,
      "learning_rate": 1.449277829877775e-06,
      "loss": 0.08969354629516602,
      "mean_token_accuracy": 0.9923582375049591,
      "num_tokens": 8532083.0,
      "step": 53
    },
    {
      "aux_loss": 8.049753189086914,
      "entropy": 0.05157726630568504,
      "epoch": 1.2891566265060241,
      "grad_norm": 0.42263299226760864,
      "learning_rate": 1.4492430435465495e-06,
      "loss": 0.09429101645946503,
      "mean_token_accuracy": 0.9907614588737488,
      "num_tokens": 8707916.0,
      "step": 54
    },
    {
      "aux_loss": 8.04571533203125,
      "entropy": 0.04540949687361717,
      "epoch": 1.3132530120481927,
      "grad_norm": 0.415887713432312,
      "learning_rate": 1.4492074398668059e-06,
      "loss": 0.09169892221689224,
      "mean_token_accuracy": 0.9909594357013702,
      "num_tokens": 8868293.0,
      "step": 55
    },
    {
      "aux_loss": 8.047487258911133,
      "entropy": 0.042821500450372696,
      "epoch": 1.3373493975903614,
      "grad_norm": 0.3941069543361664,
      "learning_rate": 1.4491710189189896e-06,
      "loss": 0.08963996917009354,
      "mean_token_accuracy": 0.9915030598640442,
      "num_tokens": 9040400.0,
      "step": 56
    },
    {
      "aux_loss": 8.046500205993652,
      "entropy": 0.04034057445824146,
      "epoch": 1.3614457831325302,
      "grad_norm": 0.3714058995246887,
      "learning_rate": 1.4491337807853928e-06,
      "loss": 0.08831468969583511,
      "mean_token_accuracy": 0.9924418330192566,
      "num_tokens": 9210005.0,
      "step": 57
    },
    {
      "aux_loss": 8.04520559310913,
      "entropy": 0.04288036748766899,
      "epoch": 1.3855421686746987,
      "grad_norm": 0.4117484986782074,
      "learning_rate": 1.4490957255501541e-06,
      "loss": 0.09062028676271439,
      "mean_token_accuracy": 0.9908726513385773,
      "num_tokens": 9354587.0,
      "step": 58
    },
    {
      "aux_loss": 8.048189163208008,
      "entropy": 0.04117992892861366,
      "epoch": 1.4096385542168675,
      "grad_norm": 0.3880350589752197,
      "learning_rate": 1.449056853299258e-06,
      "loss": 0.08878283947706223,
      "mean_token_accuracy": 0.9918811917304993,
      "num_tokens": 9530098.0,
      "step": 59
    },
    {
      "aux_loss": 8.046634674072266,
      "entropy": 0.046307843178510666,
      "epoch": 1.4337349397590362,
      "grad_norm": 0.41205379366874695,
      "learning_rate": 1.4490171641205356e-06,
      "loss": 0.0928514301776886,
      "mean_token_accuracy": 0.99020916223526,
      "num_tokens": 9696846.0,
      "step": 60
    },
    {
      "aux_loss": 8.04638147354126,
      "entropy": 0.03861385025084019,
      "epoch": 1.4578313253012047,
      "grad_norm": 0.3674647808074951,
      "learning_rate": 1.4489766581036633e-06,
      "loss": 0.08779378235340118,
      "mean_token_accuracy": 0.9924313127994537,
      "num_tokens": 9868180.0,
      "step": 61
    },
    {
      "aux_loss": 8.048831462860107,
      "entropy": 0.04485859349370003,
      "epoch": 1.4819277108433735,
      "grad_norm": 0.4122410714626312,
      "learning_rate": 1.4489353353401633e-06,
      "loss": 0.09213700890541077,
      "mean_token_accuracy": 0.9905197322368622,
      "num_tokens": 10041207.0,
      "step": 62
    },
    {
      "aux_loss": 8.045693397521973,
      "entropy": 0.04181382618844509,
      "epoch": 1.5060240963855422,
      "grad_norm": 0.41252797842025757,
      "learning_rate": 1.4488931959234036e-06,
      "loss": 0.08966240286827087,
      "mean_token_accuracy": 0.9918944835662842,
      "num_tokens": 10216601.0,
      "step": 63
    },
    {
      "aux_loss": 8.045026779174805,
      "entropy": 0.04367237165570259,
      "epoch": 1.5301204819277108,
      "grad_norm": 0.7194992899894714,
      "learning_rate": 1.4488502399485967e-06,
      "loss": 0.09064353257417679,
      "mean_token_accuracy": 0.9913205206394196,
      "num_tokens": 10373157.0,
      "step": 64
    },
    {
      "aux_loss": 8.046749591827393,
      "entropy": 0.049618642777204514,
      "epoch": 1.5542168674698795,
      "grad_norm": 0.4409625828266144,
      "learning_rate": 1.4488064675128008e-06,
      "loss": 0.09418876469135284,
      "mean_token_accuracy": 0.98995640873909,
      "num_tokens": 10513760.0,
      "step": 65
    },
    {
      "aux_loss": 8.04750108718872,
      "entropy": 0.045051392167806625,
      "epoch": 1.5783132530120483,
      "grad_norm": 0.4379954934120178,
      "learning_rate": 1.4487618787149178e-06,
      "loss": 0.09229664504528046,
      "mean_token_accuracy": 0.990358978509903,
      "num_tokens": 10667072.0,
      "step": 66
    },
    {
      "aux_loss": 8.0473952293396,
      "entropy": 0.03963223472237587,
      "epoch": 1.6024096385542168,
      "grad_norm": 0.3720748722553253,
      "learning_rate": 1.4487164736556958e-06,
      "loss": 0.08866938948631287,
      "mean_token_accuracy": 0.9920879304409027,
      "num_tokens": 10839108.0,
      "step": 67
    },
    {
      "aux_loss": 8.047708511352539,
      "entropy": 0.04451095871627331,
      "epoch": 1.6265060240963856,
      "grad_norm": 0.4250375032424927,
      "learning_rate": 1.4486702524377257e-06,
      "loss": 0.09129409492015839,
      "mean_token_accuracy": 0.9904707074165344,
      "num_tokens": 11001747.0,
      "step": 68
    },
    {
      "aux_loss": 8.045770645141602,
      "entropy": 0.041699016466736794,
      "epoch": 1.6506024096385543,
      "grad_norm": 0.5749812722206116,
      "learning_rate": 1.4486232151654433e-06,
      "loss": 0.08955338597297668,
      "mean_token_accuracy": 0.9914253950119019,
      "num_tokens": 11163010.0,
      "step": 69
    },
    {
      "aux_loss": 8.048367977142334,
      "entropy": 0.0460666548460722,
      "epoch": 1.6746987951807228,
      "grad_norm": 0.412992388010025,
      "learning_rate": 1.448575361945128e-06,
      "loss": 0.09151523560285568,
      "mean_token_accuracy": 0.9911079108715057,
      "num_tokens": 11326488.0,
      "step": 70
    },
    {
      "aux_loss": 8.047677040100098,
      "entropy": 0.039530523121356964,
      "epoch": 1.6987951807228916,
      "grad_norm": 0.3757312297821045,
      "learning_rate": 1.4485266928849028e-06,
      "loss": 0.08811481297016144,
      "mean_token_accuracy": 0.9917347729206085,
      "num_tokens": 11501286.0,
      "step": 71
    },
    {
      "aux_loss": 8.046856880187988,
      "entropy": 0.041672926396131516,
      "epoch": 1.7228915662650603,
      "grad_norm": 0.38275983929634094,
      "learning_rate": 1.448477208094734e-06,
      "loss": 0.08906584978103638,
      "mean_token_accuracy": 0.9918383359909058,
      "num_tokens": 11668188.0,
      "step": 72
    },
    {
      "aux_loss": 8.04528522491455,
      "entropy": 0.047908855602145195,
      "epoch": 1.7469879518072289,
      "grad_norm": 0.4008839726448059,
      "learning_rate": 1.4484269076864313e-06,
      "loss": 0.09260185062885284,
      "mean_token_accuracy": 0.9907302260398865,
      "num_tokens": 11834809.0,
      "step": 73
    },
    {
      "aux_loss": 8.047989845275879,
      "entropy": 0.049024928361177444,
      "epoch": 1.7710843373493976,
      "grad_norm": 0.43533122539520264,
      "learning_rate": 1.448375791773647e-06,
      "loss": 0.09409160912036896,
      "mean_token_accuracy": 0.9904188215732574,
      "num_tokens": 12004510.0,
      "step": 74
    },
    {
      "aux_loss": 8.045915126800537,
      "entropy": 0.05250510759651661,
      "epoch": 1.7951807228915664,
      "grad_norm": 0.4690086841583252,
      "learning_rate": 1.4483238604718765e-06,
      "loss": 0.0955488458275795,
      "mean_token_accuracy": 0.9906903803348541,
      "num_tokens": 12137534.0,
      "step": 75
    },
    {
      "aux_loss": 8.044984817504883,
      "entropy": 0.0376634681597352,
      "epoch": 1.819277108433735,
      "grad_norm": 0.35793742537498474,
      "learning_rate": 1.4482711138984564e-06,
      "loss": 0.08688877522945404,
      "mean_token_accuracy": 0.9927193224430084,
      "num_tokens": 12325369.0,
      "step": 76
    },
    {
      "aux_loss": 8.04779052734375,
      "entropy": 0.04571797512471676,
      "epoch": 1.8433734939759037,
      "grad_norm": 0.4300045371055603,
      "learning_rate": 1.4482175521725665e-06,
      "loss": 0.09217096120119095,
      "mean_token_accuracy": 0.9905279278755188,
      "num_tokens": 12488806.0,
      "step": 77
    },
    {
      "aux_loss": 8.044951915740967,
      "entropy": 0.05060774274170399,
      "epoch": 1.8674698795180724,
      "grad_norm": 0.423297643661499,
      "learning_rate": 1.4481631754152286e-06,
      "loss": 0.0944666936993599,
      "mean_token_accuracy": 0.990161120891571,
      "num_tokens": 12640695.0,
      "step": 78
    },
    {
      "aux_loss": 8.047436714172363,
      "entropy": 0.04260541498661041,
      "epoch": 1.891566265060241,
      "grad_norm": 0.39964279532432556,
      "learning_rate": 1.4481079837493048e-06,
      "loss": 0.08988695591688156,
      "mean_token_accuracy": 0.9916346371173859,
      "num_tokens": 12823930.0,
      "step": 79
    },
    {
      "aux_loss": 8.047076225280762,
      "entropy": 0.04447777755558491,
      "epoch": 1.9156626506024095,
      "grad_norm": 0.3796585202217102,
      "learning_rate": 1.4480519772994993e-06,
      "loss": 0.09115425497293472,
      "mean_token_accuracy": 0.9917284250259399,
      "num_tokens": 12998419.0,
      "step": 80
    },
    {
      "aux_loss": 8.04789924621582,
      "entropy": 0.04236619919538498,
      "epoch": 1.9397590361445785,
      "grad_norm": 0.418720543384552,
      "learning_rate": 1.4479951561923573e-06,
      "loss": 0.0903579443693161,
      "mean_token_accuracy": 0.9915480315685272,
      "num_tokens": 13148913.0,
      "step": 81
    },
    {
      "aux_loss": 8.045738697052002,
      "entropy": 0.04441889189183712,
      "epoch": 1.963855421686747,
      "grad_norm": 0.3862975835800171,
      "learning_rate": 1.4479375205562642e-06,
      "loss": 0.09073998779058456,
      "mean_token_accuracy": 0.9912832081317902,
      "num_tokens": 13304132.0,
      "step": 82
    },
    {
      "aux_loss": 8.046059608459473,
      "entropy": 0.05218968167901039,
      "epoch": 1.9879518072289155,
      "grad_norm": 0.4317293167114258,
      "learning_rate": 1.4478790705214464e-06,
      "loss": 0.09526760876178741,
      "mean_token_accuracy": 0.9900495111942291,
      "num_tokens": 13453623.0,
      "step": 83
    },
    {
      "aux_loss": 8.05069351196289,
      "entropy": 0.039391469210386276,
      "epoch": 2.0,
      "grad_norm": 0.6828374266624451,
      "learning_rate": 1.4478198062199706e-06,
      "loss": 0.05737518146634102,
      "mean_token_accuracy": 0.9917781352996826,
      "num_tokens": 13482926.0,
      "step": 84
    },
    {
      "aux_loss": 8.045267105102539,
      "entropy": 0.053030213341116905,
      "epoch": 2.0240963855421685,
      "grad_norm": 0.40254807472229004,
      "learning_rate": 1.4477597277857421e-06,
      "loss": 0.09399621188640594,
      "mean_token_accuracy": 0.9910189509391785,
      "num_tokens": 13616336.0,
      "step": 85
    },
    {
      "aux_loss": 8.04813003540039,
      "entropy": 0.04406110197305679,
      "epoch": 2.0481927710843375,
      "grad_norm": 0.36270391941070557,
      "learning_rate": 1.4476988353545068e-06,
      "loss": 0.08861124515533447,
      "mean_token_accuracy": 0.9931189119815826,
      "num_tokens": 13775230.0,
      "step": 86
    },
    {
      "aux_loss": 8.044662952423096,
      "entropy": 0.03888658061623573,
      "epoch": 2.072289156626506,
      "grad_norm": 0.3652919828891754,
      "learning_rate": 1.4476371290638496e-06,
      "loss": 0.08651480078697205,
      "mean_token_accuracy": 0.9931168258190155,
      "num_tokens": 13935092.0,
      "step": 87
    },
    {
      "aux_loss": 8.048104763031006,
      "entropy": 0.047589803114533424,
      "epoch": 2.0963855421686746,
      "grad_norm": 0.4187248647212982,
      "learning_rate": 1.447574609053194e-06,
      "loss": 0.09169710427522659,
      "mean_token_accuracy": 0.9922028183937073,
      "num_tokens": 14097222.0,
      "step": 88
    },
    {
      "aux_loss": 8.044262409210205,
      "entropy": 0.04133613780140877,
      "epoch": 2.1204819277108435,
      "grad_norm": 0.3797593414783478,
      "learning_rate": 1.4475112754638022e-06,
      "loss": 0.08802482485771179,
      "mean_token_accuracy": 0.9926626980304718,
      "num_tokens": 14255553.0,
      "step": 89
    },
    {
      "aux_loss": 8.046658515930176,
      "entropy": 0.04000333696603775,
      "epoch": 2.144578313253012,
      "grad_norm": 0.4277676045894623,
      "learning_rate": 1.447447128438775e-06,
      "loss": 0.08734406530857086,
      "mean_token_accuracy": 0.993060439825058,
      "num_tokens": 14431608.0,
      "step": 90
    },
    {
      "aux_loss": 8.045912265777588,
      "entropy": 0.041802020743489265,
      "epoch": 2.1686746987951806,
      "grad_norm": 0.37397268414497375,
      "learning_rate": 1.4473821681230506e-06,
      "loss": 0.08865631371736526,
      "mean_token_accuracy": 0.9919825196266174,
      "num_tokens": 14615993.0,
      "step": 91
    },
    {
      "aux_loss": 8.044597625732422,
      "entropy": 0.04039378650486469,
      "epoch": 2.1927710843373496,
      "grad_norm": 0.423145592212677,
      "learning_rate": 1.4473163946634052e-06,
      "loss": 0.0880887508392334,
      "mean_token_accuracy": 0.9921639561653137,
      "num_tokens": 14766575.0,
      "step": 92
    },
    {
      "aux_loss": 8.04832410812378,
      "entropy": 0.04441938176751137,
      "epoch": 2.216867469879518,
      "grad_norm": 0.3814225196838379,
      "learning_rate": 1.4472498082084518e-06,
      "loss": 0.08906400948762894,
      "mean_token_accuracy": 0.9928748905658722,
      "num_tokens": 14931445.0,
      "step": 93
    },
    {
      "aux_loss": 8.042942523956299,
      "entropy": 0.046976158395409584,
      "epoch": 2.2409638554216866,
      "grad_norm": 0.3922480344772339,
      "learning_rate": 1.4471824089086415e-06,
      "loss": 0.09150424599647522,
      "mean_token_accuracy": 0.9916513860225677,
      "num_tokens": 15091210.0,
      "step": 94
    },
    {
      "aux_loss": 8.045757293701172,
      "entropy": 0.035969141870737076,
      "epoch": 2.2650602409638556,
      "grad_norm": 0.3510456681251526,
      "learning_rate": 1.44711419691626e-06,
      "loss": 0.0854613184928894,
      "mean_token_accuracy": 0.9934249818325043,
      "num_tokens": 15260222.0,
      "step": 95
    },
    {
      "aux_loss": 8.044999122619629,
      "entropy": 0.03909716196358204,
      "epoch": 2.289156626506024,
      "grad_norm": 0.3629781901836395,
      "learning_rate": 1.4470451723854314e-06,
      "loss": 0.08637961745262146,
      "mean_token_accuracy": 0.9935233891010284,
      "num_tokens": 15455860.0,
      "step": 96
    },
    {
      "aux_loss": 8.04701280593872,
      "entropy": 0.04057766683399677,
      "epoch": 2.3132530120481927,
      "grad_norm": 0.36879613995552063,
      "learning_rate": 1.4469753354721144e-06,
      "loss": 0.087168388068676,
      "mean_token_accuracy": 0.9932206869125366,
      "num_tokens": 15631021.0,
      "step": 97
    },
    {
      "aux_loss": 8.047924518585205,
      "entropy": 0.04564652778208256,
      "epoch": 2.337349397590361,
      "grad_norm": 0.41153064370155334,
      "learning_rate": 1.4469046863341036e-06,
      "loss": 0.08925162255764008,
      "mean_token_accuracy": 0.9921718537807465,
      "num_tokens": 15789136.0,
      "step": 98
    },
    {
      "aux_loss": 8.0452880859375,
      "entropy": 0.04301636107265949,
      "epoch": 2.36144578313253,
      "grad_norm": 0.376197874546051,
      "learning_rate": 1.4468332251310286e-06,
      "loss": 0.08848991990089417,
      "mean_token_accuracy": 0.9930213391780853,
      "num_tokens": 15938397.0,
      "step": 99
    },
    {
      "aux_loss": 8.043729782104492,
      "entropy": 0.04113657400012016,
      "epoch": 2.3855421686746987,
      "grad_norm": 0.3817802667617798,
      "learning_rate": 1.4467609520243544e-06,
      "loss": 0.08791770040988922,
      "mean_token_accuracy": 0.9926939308643341,
      "num_tokens": 16114803.0,
      "step": 100
    },
    {
      "aux_loss": 8.044564247131348,
      "entropy": 0.04448202811181545,
      "epoch": 2.4096385542168672,
      "grad_norm": 0.41418957710266113,
      "learning_rate": 1.44668786717738e-06,
      "loss": 0.08933436870574951,
      "mean_token_accuracy": 0.9923655092716217,
      "num_tokens": 16267681.0,
      "step": 101
    },
    {
      "aux_loss": 8.045689105987549,
      "entropy": 0.043511977419257164,
      "epoch": 2.433734939759036,
      "grad_norm": 0.3912562429904938,
      "learning_rate": 1.4466139707552382e-06,
      "loss": 0.08852104097604752,
      "mean_token_accuracy": 0.9933134019374847,
      "num_tokens": 16435524.0,
      "step": 102
    },
    {
      "aux_loss": 8.04603910446167,
      "entropy": 0.03945927880704403,
      "epoch": 2.4578313253012047,
      "grad_norm": 0.37696266174316406,
      "learning_rate": 1.4465392629248964e-06,
      "loss": 0.08802925050258636,
      "mean_token_accuracy": 0.9925020635128021,
      "num_tokens": 16601843.0,
      "step": 103
    },
    {
      "aux_loss": 8.041633129119873,
      "entropy": 0.042852023616433144,
      "epoch": 2.4819277108433733,
      "grad_norm": 0.42523500323295593,
      "learning_rate": 1.446463743855155e-06,
      "loss": 0.08881749212741852,
      "mean_token_accuracy": 0.9926598370075226,
      "num_tokens": 16749225.0,
      "step": 104
    },
    {
      "aux_loss": 8.044000625610352,
      "entropy": 0.04509095661342144,
      "epoch": 2.5060240963855422,
      "grad_norm": 0.43666303157806396,
      "learning_rate": 1.4463874137166468e-06,
      "loss": 0.09076739847660065,
      "mean_token_accuracy": 0.9915396869182587,
      "num_tokens": 16897018.0,
      "step": 105
    },
    {
      "aux_loss": 8.044254302978516,
      "entropy": 0.04391391761600971,
      "epoch": 2.5301204819277108,
      "grad_norm": 0.4111824333667755,
      "learning_rate": 1.4463102726818382e-06,
      "loss": 0.08995572477579117,
      "mean_token_accuracy": 0.9917828142642975,
      "num_tokens": 17065753.0,
      "step": 106
    },
    {
      "aux_loss": 8.045661926269531,
      "entropy": 0.04082946851849556,
      "epoch": 2.5542168674698793,
      "grad_norm": 0.3910435438156128,
      "learning_rate": 1.4462323209250268e-06,
      "loss": 0.08806779235601425,
      "mean_token_accuracy": 0.9928287863731384,
      "num_tokens": 17235799.0,
      "step": 107
    },
    {
      "aux_loss": 8.043983459472656,
      "entropy": 0.045222554355859756,
      "epoch": 2.5783132530120483,
      "grad_norm": 0.4386132061481476,
      "learning_rate": 1.4461535586223427e-06,
      "loss": 0.090940460562706,
      "mean_token_accuracy": 0.9915510416030884,
      "num_tokens": 17376992.0,
      "step": 108
    },
    {
      "aux_loss": 8.0429105758667,
      "entropy": 0.04239073768258095,
      "epoch": 2.602409638554217,
      "grad_norm": 0.39754438400268555,
      "learning_rate": 1.4460739859517472e-06,
      "loss": 0.08899098634719849,
      "mean_token_accuracy": 0.9922184944152832,
      "num_tokens": 17537095.0,
      "step": 109
    },
    {
      "aux_loss": 8.046865940093994,
      "entropy": 0.04032203182578087,
      "epoch": 2.6265060240963853,
      "grad_norm": 0.4198264181613922,
      "learning_rate": 1.4459936030930326e-06,
      "loss": 0.08784371614456177,
      "mean_token_accuracy": 0.9927988350391388,
      "num_tokens": 17697756.0,
      "step": 110
    },
    {
      "aux_loss": 8.045022964477539,
      "entropy": 0.04641672037541866,
      "epoch": 2.6506024096385543,
      "grad_norm": 0.41483283042907715,
      "learning_rate": 1.4459124102278213e-06,
      "loss": 0.0906679630279541,
      "mean_token_accuracy": 0.9918768405914307,
      "num_tokens": 17862032.0,
      "step": 111
    },
    {
      "aux_loss": 8.046215057373047,
      "entropy": 0.038604725152254105,
      "epoch": 2.674698795180723,
      "grad_norm": 0.38772159814834595,
      "learning_rate": 1.4458304075395668e-06,
      "loss": 0.08620807528495789,
      "mean_token_accuracy": 0.9930865466594696,
      "num_tokens": 18014648.0,
      "step": 112
    },
    {
      "aux_loss": 8.04534912109375,
      "entropy": 0.047094520181417465,
      "epoch": 2.6987951807228914,
      "grad_norm": 0.4265570342540741,
      "learning_rate": 1.445747595213552e-06,
      "loss": 0.09142982959747314,
      "mean_token_accuracy": 0.9916557669639587,
      "num_tokens": 18170137.0,
      "step": 113
    },
    {
      "aux_loss": 8.045645713806152,
      "entropy": 0.037863900884985924,
      "epoch": 2.7228915662650603,
      "grad_norm": 0.3691053092479706,
      "learning_rate": 1.4456639734368888e-06,
      "loss": 0.08656749129295349,
      "mean_token_accuracy": 0.9931949079036713,
      "num_tokens": 18341196.0,
      "step": 114
    },
    {
      "aux_loss": 8.04396677017212,
      "entropy": 0.04190782643854618,
      "epoch": 2.746987951807229,
      "grad_norm": 0.38775166869163513,
      "learning_rate": 1.4455795423985187e-06,
      "loss": 0.08832530677318573,
      "mean_token_accuracy": 0.9925132691860199,
      "num_tokens": 18499534.0,
      "step": 115
    },
    {
      "aux_loss": 8.046361446380615,
      "entropy": 0.044008735567331314,
      "epoch": 2.7710843373493974,
      "grad_norm": 0.3934473395347595,
      "learning_rate": 1.445494302289211e-06,
      "loss": 0.08926442265510559,
      "mean_token_accuracy": 0.9925075769424438,
      "num_tokens": 18659760.0,
      "step": 116
    },
    {
      "aux_loss": 8.044501304626465,
      "entropy": 0.04429285787045956,
      "epoch": 2.7951807228915664,
      "grad_norm": 0.4424228072166443,
      "learning_rate": 1.4454082533015636e-06,
      "loss": 0.08979617059230804,
      "mean_token_accuracy": 0.9921565651893616,
      "num_tokens": 18831165.0,
      "step": 117
    },
    {
      "aux_loss": 8.042979717254639,
      "entropy": 0.040504178032279015,
      "epoch": 2.819277108433735,
      "grad_norm": 0.4018898904323578,
      "learning_rate": 1.4453213956300015e-06,
      "loss": 0.08786608278751373,
      "mean_token_accuracy": 0.992399662733078,
      "num_tokens": 18984127.0,
      "step": 118
    },
    {
      "aux_loss": 8.045685768127441,
      "entropy": 0.041042402386665344,
      "epoch": 2.8433734939759034,
      "grad_norm": 0.5288944840431213,
      "learning_rate": 1.4452337294707773e-06,
      "loss": 0.08859241008758545,
      "mean_token_accuracy": 0.9921814501285553,
      "num_tokens": 19160419.0,
      "step": 119
    },
    {
      "aux_loss": 8.046215057373047,
      "entropy": 0.04414335638284683,
      "epoch": 2.8674698795180724,
      "grad_norm": 0.4066352844238281,
      "learning_rate": 1.4451452550219706e-06,
      "loss": 0.08978652209043503,
      "mean_token_accuracy": 0.9923945069313049,
      "num_tokens": 19335546.0,
      "step": 120
    },
    {
      "aux_loss": 8.044774055480957,
      "entropy": 0.03940526768565178,
      "epoch": 2.891566265060241,
      "grad_norm": 0.4042900800704956,
      "learning_rate": 1.445055972483487e-06,
      "loss": 0.08746181428432465,
      "mean_token_accuracy": 0.9926989078521729,
      "num_tokens": 19496314.0,
      "step": 121
    },
    {
      "aux_loss": 8.043519020080566,
      "entropy": 0.03671685978770256,
      "epoch": 2.9156626506024095,
      "grad_norm": 0.3954128921031952,
      "learning_rate": 1.4449658820570577e-06,
      "loss": 0.08653290569782257,
      "mean_token_accuracy": 0.9926026463508606,
      "num_tokens": 19661718.0,
      "step": 122
    },
    {
      "aux_loss": 8.045527458190918,
      "entropy": 0.04113632068037987,
      "epoch": 2.9397590361445785,
      "grad_norm": 0.39803871512413025,
      "learning_rate": 1.4448749839462395e-06,
      "loss": 0.08833590149879456,
      "mean_token_accuracy": 0.9923411309719086,
      "num_tokens": 19847373.0,
      "step": 123
    },
    {
      "aux_loss": 8.044186115264893,
      "entropy": 0.03642796538770199,
      "epoch": 2.963855421686747,
      "grad_norm": 0.3793511390686035,
      "learning_rate": 1.4447832783564146e-06,
      "loss": 0.086070716381073,
      "mean_token_accuracy": 0.9930998384952545,
      "num_tokens": 20025875.0,
      "step": 124
    },
    {
      "aux_loss": 8.046002388000488,
      "entropy": 0.047230470925569534,
      "epoch": 2.9879518072289155,
      "grad_norm": 0.4477235972881317,
      "learning_rate": 1.444690765494789e-06,
      "loss": 0.09117428958415985,
      "mean_token_accuracy": 0.9920151233673096,
      "num_tokens": 20182590.0,
      "step": 125
    },
    {
      "aux_loss": 8.042522430419922,
      "entropy": 0.040220800787210464,
      "epoch": 3.0,
      "grad_norm": 0.5846906900405884,
      "learning_rate": 1.4445974455703927e-06,
      "loss": 0.0573921799659729,
      "mean_token_accuracy": 0.9929695129394531,
      "num_tokens": 20224389.0,
      "step": 126
    },
    {
      "aux_loss": 8.044944763183594,
      "entropy": 0.039370276033878326,
      "epoch": 3.0240963855421685,
      "grad_norm": 0.3742491900920868,
      "learning_rate": 1.44450331879408e-06,
      "loss": 0.08556805551052094,
      "mean_token_accuracy": 0.9942886531352997,
      "num_tokens": 20380640.0,
      "step": 127
    },
    {
      "aux_loss": 8.043789863586426,
      "entropy": 0.045225948095321655,
      "epoch": 3.0481927710843375,
      "grad_norm": 0.38408586382865906,
      "learning_rate": 1.4444083853785277e-06,
      "loss": 0.08862867951393127,
      "mean_token_accuracy": 0.9931619465351105,
      "num_tokens": 20532981.0,
      "step": 128
    },
    {
      "aux_loss": 8.042508125305176,
      "entropy": 0.0450060423463583,
      "epoch": 3.072289156626506,
      "grad_norm": 0.37083542346954346,
      "learning_rate": 1.444312645538235e-06,
      "loss": 0.08864334225654602,
      "mean_token_accuracy": 0.9928875267505646,
      "num_tokens": 20683874.0,
      "step": 129
    },
    {
      "aux_loss": 8.041747093200684,
      "entropy": 0.04171221703290939,
      "epoch": 3.0963855421686746,
      "grad_norm": 0.36446937918663025,
      "learning_rate": 1.4442160994895236e-06,
      "loss": 0.08605008572340012,
      "mean_token_accuracy": 0.9937500953674316,
      "num_tokens": 20836727.0,
      "step": 130
    },
    {
      "aux_loss": 8.042478561401367,
      "entropy": 0.04033582843840122,
      "epoch": 3.1204819277108435,
      "grad_norm": 0.37039053440093994,
      "learning_rate": 1.4441187474505368e-06,
      "loss": 0.08619973808526993,
      "mean_token_accuracy": 0.9936582446098328,
      "num_tokens": 20989173.0,
      "step": 131
    },
    {
      "aux_loss": 8.04190444946289,
      "entropy": 0.04400304704904556,
      "epoch": 3.144578313253012,
      "grad_norm": 0.39441367983818054,
      "learning_rate": 1.4440205896412386e-06,
      "loss": 0.08861161768436432,
      "mean_token_accuracy": 0.993509829044342,
      "num_tokens": 21137746.0,
      "step": 132
    },
    {
      "aux_loss": 8.042577743530273,
      "entropy": 0.0362233929336071,
      "epoch": 3.1686746987951806,
      "grad_norm": 0.3513298034667969,
      "learning_rate": 1.4439216262834142e-06,
      "loss": 0.08395665884017944,
      "mean_token_accuracy": 0.9942896068096161,
      "num_tokens": 21303083.0,
      "step": 133
    },
    {
      "aux_loss": 8.045822143554688,
      "entropy": 0.044513555243611336,
      "epoch": 3.1927710843373496,
      "grad_norm": 0.4062028229236603,
      "learning_rate": 1.4438218576006684e-06,
      "loss": 0.08783506602048874,
      "mean_token_accuracy": 0.9932784140110016,
      "num_tokens": 21465720.0,
      "step": 134
    },
    {
      "aux_loss": 8.042749881744385,
      "entropy": 0.03609209880232811,
      "epoch": 3.216867469879518,
      "grad_norm": 0.36185288429260254,
      "learning_rate": 1.4437212838184255e-06,
      "loss": 0.08398795872926712,
      "mean_token_accuracy": 0.9942994117736816,
      "num_tokens": 21634832.0,
      "step": 135
    },
    {
      "aux_loss": 8.046396255493164,
      "entropy": 0.039127200841903687,
      "epoch": 3.2409638554216866,
      "grad_norm": 0.3764151334762573,
      "learning_rate": 1.44361990516393e-06,
      "loss": 0.08616587519645691,
      "mean_token_accuracy": 0.993415892124176,
      "num_tokens": 21801355.0,
      "step": 136
    },
    {
      "aux_loss": 8.043558597564697,
      "entropy": 0.03838344104588032,
      "epoch": 3.2650602409638556,
      "grad_norm": 0.3657795786857605,
      "learning_rate": 1.4435177218662435e-06,
      "loss": 0.08551836758852005,
      "mean_token_accuracy": 0.9938235878944397,
      "num_tokens": 21973015.0,
      "step": 137
    },
    {
      "aux_loss": 8.043213367462158,
      "entropy": 0.04633874632418156,
      "epoch": 3.289156626506024,
      "grad_norm": 0.38201189041137695,
      "learning_rate": 1.4434147341562468e-06,
      "loss": 0.08937641978263855,
      "mean_token_accuracy": 0.9929512739181519,
      "num_tokens": 22134914.0,
      "step": 138
    },
    {
      "aux_loss": 8.04480266571045,
      "entropy": 0.03689238056540489,
      "epoch": 3.3132530120481927,
      "grad_norm": 0.35625335574150085,
      "learning_rate": 1.4433109422666374e-06,
      "loss": 0.08385109901428223,
      "mean_token_accuracy": 0.9944583773612976,
      "num_tokens": 22302240.0,
      "step": 139
    },
    {
      "aux_loss": 8.042608261108398,
      "entropy": 0.04020004719495773,
      "epoch": 3.337349397590361,
      "grad_norm": 0.3919045925140381,
      "learning_rate": 1.4432063464319306e-06,
      "loss": 0.0874231606721878,
      "mean_token_accuracy": 0.9931889474391937,
      "num_tokens": 22471698.0,
      "step": 140
    },
    {
      "aux_loss": 8.042847156524658,
      "entropy": 0.03685566782951355,
      "epoch": 3.36144578313253,
      "grad_norm": 0.3515630066394806,
      "learning_rate": 1.4431009468884578e-06,
      "loss": 0.0849953442811966,
      "mean_token_accuracy": 0.9941838085651398,
      "num_tokens": 22653073.0,
      "step": 141
    },
    {
      "aux_loss": 8.043398380279541,
      "entropy": 0.0423385389149189,
      "epoch": 3.3855421686746987,
      "grad_norm": 0.41736137866973877,
      "learning_rate": 1.442994743874366e-06,
      "loss": 0.08791652321815491,
      "mean_token_accuracy": 0.9927142858505249,
      "num_tokens": 22810578.0,
      "step": 142
    },
    {
      "aux_loss": 8.04459524154663,
      "entropy": 0.04449957422912121,
      "epoch": 3.4096385542168672,
      "grad_norm": 0.4232349693775177,
      "learning_rate": 1.4428877376296185e-06,
      "loss": 0.08924206346273422,
      "mean_token_accuracy": 0.9924839735031128,
      "num_tokens": 22953794.0,
      "step": 143
    },
    {
      "aux_loss": 8.041166305541992,
      "entropy": 0.04270816035568714,
      "epoch": 3.433734939759036,
      "grad_norm": 0.3914850354194641,
      "learning_rate": 1.4427799283959926e-06,
      "loss": 0.08785770833492279,
      "mean_token_accuracy": 0.9931382238864899,
      "num_tokens": 23120357.0,
      "step": 144
    },
    {
      "aux_loss": 8.04544448852539,
      "entropy": 0.03356307931244373,
      "epoch": 3.4578313253012047,
      "grad_norm": 0.3500516712665558,
      "learning_rate": 1.4426713164170808e-06,
      "loss": 0.08334299921989441,
      "mean_token_accuracy": 0.9946193993091583,
      "num_tokens": 23300231.0,
      "step": 145
    },
    {
      "aux_loss": 8.04530382156372,
      "entropy": 0.044576091691851616,
      "epoch": 3.4819277108433733,
      "grad_norm": 0.41992995142936707,
      "learning_rate": 1.4425619019382884e-06,
      "loss": 0.08873672783374786,
      "mean_token_accuracy": 0.9929416477680206,
      "num_tokens": 23453268.0,
      "step": 146
    },
    {
      "aux_loss": 8.044836044311523,
      "entropy": 0.03937925212085247,
      "epoch": 3.5060240963855422,
      "grad_norm": 0.37806037068367004,
      "learning_rate": 1.4424516852068346e-06,
      "loss": 0.08648265898227692,
      "mean_token_accuracy": 0.9935058355331421,
      "num_tokens": 23625257.0,
      "step": 147
    },
    {
      "aux_loss": 8.04463529586792,
      "entropy": 0.0399188157171011,
      "epoch": 3.5301204819277108,
      "grad_norm": 0.37204214930534363,
      "learning_rate": 1.442340666471751e-06,
      "loss": 0.08639834821224213,
      "mean_token_accuracy": 0.9933888912200928,
      "num_tokens": 23820327.0,
      "step": 148
    },
    {
      "aux_loss": 8.044984340667725,
      "entropy": 0.036087172105908394,
      "epoch": 3.5542168674698793,
      "grad_norm": 0.36808425188064575,
      "learning_rate": 1.4422288459838816e-06,
      "loss": 0.08478154242038727,
      "mean_token_accuracy": 0.9945113360881805,
      "num_tokens": 23979995.0,
      "step": 149
    },
    {
      "aux_loss": 8.043323993682861,
      "entropy": 0.04419667460024357,
      "epoch": 3.5783132530120483,
      "grad_norm": 0.41120970249176025,
      "learning_rate": 1.4421162239958818e-06,
      "loss": 0.0894618034362793,
      "mean_token_accuracy": 0.9925802052021027,
      "num_tokens": 24125572.0,
      "step": 150
    },
    {
      "aux_loss": 8.042481422424316,
      "entropy": 0.03813181072473526,
      "epoch": 3.602409638554217,
      "grad_norm": 0.3920314610004425,
      "learning_rate": 1.4420028007622175e-06,
      "loss": 0.08617769181728363,
      "mean_token_accuracy": 0.9939529597759247,
      "num_tokens": 24283705.0,
      "step": 151
    },
    {
      "aux_loss": 8.043877601623535,
      "entropy": 0.03504548221826553,
      "epoch": 3.6265060240963853,
      "grad_norm": 0.38018906116485596,
      "learning_rate": 1.4418885765391659e-06,
      "loss": 0.08467784523963928,
      "mean_token_accuracy": 0.993707150220871,
      "num_tokens": 24472505.0,
      "step": 152
    },
    {
      "aux_loss": 8.042401313781738,
      "entropy": 0.04122787527740002,
      "epoch": 3.6506024096385543,
      "grad_norm": 0.39765456318855286,
      "learning_rate": 1.4417735515848131e-06,
      "loss": 0.08704474568367004,
      "mean_token_accuracy": 0.9933643937110901,
      "num_tokens": 24643360.0,
      "step": 153
    },
    {
      "aux_loss": 8.043309688568115,
      "entropy": 0.04166724532842636,
      "epoch": 3.674698795180723,
      "grad_norm": 0.3905385136604309,
      "learning_rate": 1.4416577261590549e-06,
      "loss": 0.08706511557102203,
      "mean_token_accuracy": 0.9933055341243744,
      "num_tokens": 24802344.0,
      "step": 154
    },
    {
      "aux_loss": 8.040821552276611,
      "entropy": 0.03833147883415222,
      "epoch": 3.6987951807228914,
      "grad_norm": 0.4488506615161896,
      "learning_rate": 1.4415411005235958e-06,
      "loss": 0.08571713417768478,
      "mean_token_accuracy": 0.9934513568878174,
      "num_tokens": 24956387.0,
      "step": 155
    },
    {
      "aux_loss": 8.043232440948486,
      "entropy": 0.03610091283917427,
      "epoch": 3.7228915662650603,
      "grad_norm": 0.3729538023471832,
      "learning_rate": 1.4414236749419483e-06,
      "loss": 0.08453528583049774,
      "mean_token_accuracy": 0.9941688776016235,
      "num_tokens": 25118460.0,
      "step": 156
    },
    {
      "aux_loss": 8.042871475219727,
      "entropy": 0.04386945813894272,
      "epoch": 3.746987951807229,
      "grad_norm": 0.4113171100616455,
      "learning_rate": 1.441305449679432e-06,
      "loss": 0.08860649913549423,
      "mean_token_accuracy": 0.9927978813648224,
      "num_tokens": 25276453.0,
      "step": 157
    },
    {
      "aux_loss": 8.04326343536377,
      "entropy": 0.03775089047849178,
      "epoch": 3.7710843373493974,
      "grad_norm": 0.3969726860523224,
      "learning_rate": 1.441186425003174e-06,
      "loss": 0.08585880696773529,
      "mean_token_accuracy": 0.9934997260570526,
      "num_tokens": 25465189.0,
      "step": 158
    },
    {
      "aux_loss": 8.041276931762695,
      "entropy": 0.03554354049265385,
      "epoch": 3.7951807228915664,
      "grad_norm": 0.37212640047073364,
      "learning_rate": 1.4410666011821071e-06,
      "loss": 0.08398972451686859,
      "mean_token_accuracy": 0.9942491948604584,
      "num_tokens": 25647091.0,
      "step": 159
    },
    {
      "aux_loss": 8.04280948638916,
      "entropy": 0.03362584114074707,
      "epoch": 3.819277108433735,
      "grad_norm": 0.37804487347602844,
      "learning_rate": 1.44094597848697e-06,
      "loss": 0.08355149626731873,
      "mean_token_accuracy": 0.9944404661655426,
      "num_tokens": 25812959.0,
      "step": 160
    },
    {
      "aux_loss": 8.041194915771484,
      "entropy": 0.04549684002995491,
      "epoch": 3.8433734939759034,
      "grad_norm": 0.4044184982776642,
      "learning_rate": 1.4408245571903066e-06,
      "loss": 0.08932945877313614,
      "mean_token_accuracy": 0.9926709532737732,
      "num_tokens": 25975077.0,
      "step": 161
    },
    {
      "aux_loss": 8.040953636169434,
      "entropy": 0.04158872179687023,
      "epoch": 3.8674698795180724,
      "grad_norm": 0.4143630564212799,
      "learning_rate": 1.4407023375664644e-06,
      "loss": 0.08703690767288208,
      "mean_token_accuracy": 0.993314117193222,
      "num_tokens": 26133481.0,
      "step": 162
    },
    {
      "aux_loss": 8.042658805847168,
      "entropy": 0.04513395018875599,
      "epoch": 3.891566265060241,
      "grad_norm": 0.4163261651992798,
      "learning_rate": 1.440579319891596e-06,
      "loss": 0.08887109160423279,
      "mean_token_accuracy": 0.9932331144809723,
      "num_tokens": 26285677.0,
      "step": 163
    },
    {
      "aux_loss": 8.044297218322754,
      "entropy": 0.03768927603960037,
      "epoch": 3.9156626506024095,
      "grad_norm": 0.3762443959712982,
      "learning_rate": 1.440455504443656e-06,
      "loss": 0.08547879010438919,
      "mean_token_accuracy": 0.9941076934337616,
      "num_tokens": 26456101.0,
      "step": 164
    },
    {
      "aux_loss": 8.040573596954346,
      "entropy": 0.040223028510808945,
      "epoch": 3.9397590361445785,
      "grad_norm": 0.44806984066963196,
      "learning_rate": 1.4403308915024023e-06,
      "loss": 0.08660810440778732,
      "mean_token_accuracy": 0.9934520721435547,
      "num_tokens": 26599076.0,
      "step": 165
    },
    {
      "aux_loss": 8.042471408843994,
      "entropy": 0.038684969767928123,
      "epoch": 3.963855421686747,
      "grad_norm": 0.40352219343185425,
      "learning_rate": 1.4402054813493942e-06,
      "loss": 0.08614273369312286,
      "mean_token_accuracy": 0.9937964081764221,
      "num_tokens": 26769500.0,
      "step": 166
    },
    {
      "aux_loss": 8.041973114013672,
      "entropy": 0.037965722382068634,
      "epoch": 3.9879518072289155,
      "grad_norm": 0.38133445382118225,
      "learning_rate": 1.4400792742679925e-06,
      "loss": 0.08504779636859894,
      "mean_token_accuracy": 0.9941068291664124,
      "num_tokens": 26942979.0,
      "step": 167
    },
    {
      "aux_loss": 8.039653778076172,
      "entropy": 0.046194545924663544,
      "epoch": 4.0,
      "grad_norm": 0.5609819889068604,
      "learning_rate": 1.439952270543359e-06,
      "loss": 0.053759872913360596,
      "mean_token_accuracy": 0.9936670064926147,
      "num_tokens": 26965852.0,
      "step": 168
    }
  ],
  "logging_steps": 1,
  "max_steps": 2100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "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": 5.58725343273078e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}