{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984152139461173,
  "eval_steps": 500,
  "global_step": 630,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00792393026941363,
      "grad_norm": 6.163800239562988,
      "learning_rate": 4.324992481203008e-06,
      "loss": 0.3713,
      "mean_token_accuracy": 0.8858542323112488,
      "num_tokens": 392156.0,
      "step": 5
    },
    {
      "epoch": 0.01584786053882726,
      "grad_norm": 2.714646339416504,
      "learning_rate": 9.731233082706766e-06,
      "loss": 0.2071,
      "mean_token_accuracy": 0.946096396446228,
      "num_tokens": 777999.0,
      "step": 10
    },
    {
      "epoch": 0.02377179080824089,
      "grad_norm": 1.7288715839385986,
      "learning_rate": 1.513747368421053e-05,
      "loss": 0.1566,
      "mean_token_accuracy": 0.9528929412364959,
      "num_tokens": 1178252.0,
      "step": 15
    },
    {
      "epoch": 0.03169572107765452,
      "grad_norm": 0.9263519048690796,
      "learning_rate": 2.0543714285714285e-05,
      "loss": 0.1316,
      "mean_token_accuracy": 0.962179285287857,
      "num_tokens": 1571282.0,
      "step": 20
    },
    {
      "epoch": 0.039619651347068144,
      "grad_norm": 1.1445379257202148,
      "learning_rate": 2.594995488721805e-05,
      "loss": 0.1167,
      "mean_token_accuracy": 0.9656470894813538,
      "num_tokens": 1972011.0,
      "step": 25
    },
    {
      "epoch": 0.04754358161648178,
      "grad_norm": 0.9169177412986755,
      "learning_rate": 3.135619548872181e-05,
      "loss": 0.1222,
      "mean_token_accuracy": 0.9652595818042755,
      "num_tokens": 2365035.0,
      "step": 30
    },
    {
      "epoch": 0.0554675118858954,
      "grad_norm": 2.64329195022583,
      "learning_rate": 3.676243609022557e-05,
      "loss": 0.1666,
      "mean_token_accuracy": 0.9535817563533783,
      "num_tokens": 2755914.0,
      "step": 35
    },
    {
      "epoch": 0.06339144215530904,
      "grad_norm": 0.6630797982215881,
      "learning_rate": 3.784051927887756e-05,
      "loss": 0.1418,
      "mean_token_accuracy": 0.9604487955570221,
      "num_tokens": 3149876.0,
      "step": 40
    },
    {
      "epoch": 0.07131537242472266,
      "grad_norm": 0.9154858589172363,
      "learning_rate": 3.782766416343561e-05,
      "loss": 0.1377,
      "mean_token_accuracy": 0.9591774344444275,
      "num_tokens": 3533030.0,
      "step": 45
    },
    {
      "epoch": 0.07923930269413629,
      "grad_norm": 1.018001675605774,
      "learning_rate": 3.780493000801809e-05,
      "loss": 0.1381,
      "mean_token_accuracy": 0.9634034812450409,
      "num_tokens": 3925808.0,
      "step": 50
    },
    {
      "epoch": 0.08716323296354993,
      "grad_norm": 1.2116237878799438,
      "learning_rate": 3.7772332656427635e-05,
      "loss": 0.1524,
      "mean_token_accuracy": 0.9577588558197021,
      "num_tokens": 4315464.0,
      "step": 55
    },
    {
      "epoch": 0.09508716323296355,
      "grad_norm": 0.8818077445030212,
      "learning_rate": 3.772989482628866e-05,
      "loss": 0.1126,
      "mean_token_accuracy": 0.9638817489147187,
      "num_tokens": 4712632.0,
      "step": 60
    },
    {
      "epoch": 0.10301109350237718,
      "grad_norm": 0.9603568911552429,
      "learning_rate": 3.767764609321513e-05,
      "loss": 0.1178,
      "mean_token_accuracy": 0.965680468082428,
      "num_tokens": 5096125.0,
      "step": 65
    },
    {
      "epoch": 0.1109350237717908,
      "grad_norm": 1.0121697187423706,
      "learning_rate": 3.761562287019881e-05,
      "loss": 0.1199,
      "mean_token_accuracy": 0.9638425707817078,
      "num_tokens": 5486421.0,
      "step": 70
    },
    {
      "epoch": 0.11885895404120443,
      "grad_norm": 1.0224254131317139,
      "learning_rate": 3.754386838223246e-05,
      "loss": 0.1229,
      "mean_token_accuracy": 0.9631611824035644,
      "num_tokens": 5874026.0,
      "step": 75
    },
    {
      "epoch": 0.12678288431061807,
      "grad_norm": 0.9273995161056519,
      "learning_rate": 3.746243263618564e-05,
      "loss": 0.1124,
      "mean_token_accuracy": 0.9684963762760163,
      "num_tokens": 6261499.0,
      "step": 80
    },
    {
      "epoch": 0.1347068145800317,
      "grad_norm": 0.9361715316772461,
      "learning_rate": 3.7371372385954096e-05,
      "loss": 0.1702,
      "mean_token_accuracy": 0.9554619431495667,
      "num_tokens": 6650742.0,
      "step": 85
    },
    {
      "epoch": 0.14263074484944532,
      "grad_norm": 0.7577023506164551,
      "learning_rate": 3.727075109290706e-05,
      "loss": 0.1261,
      "mean_token_accuracy": 0.9629237949848175,
      "num_tokens": 7039472.0,
      "step": 90
    },
    {
      "epoch": 0.15055467511885895,
      "grad_norm": 1.4060338735580444,
      "learning_rate": 3.7160638881660015e-05,
      "loss": 0.148,
      "mean_token_accuracy": 0.956637841463089,
      "num_tokens": 7429726.0,
      "step": 95
    },
    {
      "epoch": 0.15847860538827258,
      "grad_norm": 0.7075062394142151,
      "learning_rate": 3.7041112491203635e-05,
      "loss": 0.1312,
      "mean_token_accuracy": 0.9635329186916352,
      "num_tokens": 7817560.0,
      "step": 100
    },
    {
      "epoch": 0.1664025356576862,
      "grad_norm": 0.7909790277481079,
      "learning_rate": 3.691225522142321e-05,
      "loss": 0.1248,
      "mean_token_accuracy": 0.9630934000015259,
      "num_tokens": 8211804.0,
      "step": 105
    },
    {
      "epoch": 0.17432646592709986,
      "grad_norm": 0.8224499225616455,
      "learning_rate": 3.677415687504555e-05,
      "loss": 0.1417,
      "mean_token_accuracy": 0.9576405048370361,
      "num_tokens": 8608530.0,
      "step": 110
    },
    {
      "epoch": 0.18225039619651348,
      "grad_norm": 0.5363962054252625,
      "learning_rate": 3.6626913695054076e-05,
      "loss": 0.1231,
      "mean_token_accuracy": 0.9651646792888642,
      "num_tokens": 9004456.0,
      "step": 115
    },
    {
      "epoch": 0.1901743264659271,
      "grad_norm": 0.6564658284187317,
      "learning_rate": 3.6470628297615407e-05,
      "loss": 0.1244,
      "mean_token_accuracy": 0.9647009611129761,
      "num_tokens": 9396457.0,
      "step": 120
    },
    {
      "epoch": 0.19809825673534073,
      "grad_norm": 0.7902948260307312,
      "learning_rate": 3.630540960056459e-05,
      "loss": 0.1342,
      "mean_token_accuracy": 0.9618301033973694,
      "num_tokens": 9772407.0,
      "step": 125
    },
    {
      "epoch": 0.20602218700475436,
      "grad_norm": 0.7602660059928894,
      "learning_rate": 3.613137274749844e-05,
      "loss": 0.1114,
      "mean_token_accuracy": 0.9682710647583008,
      "num_tokens": 10158011.0,
      "step": 130
    },
    {
      "epoch": 0.21394611727416799,
      "grad_norm": 0.8078103065490723,
      "learning_rate": 3.594863902753009e-05,
      "loss": 0.1421,
      "mean_token_accuracy": 0.9573643863201141,
      "num_tokens": 10543524.0,
      "step": 135
    },
    {
      "epoch": 0.2218700475435816,
      "grad_norm": 0.6968039870262146,
      "learning_rate": 3.575733579076067e-05,
      "loss": 0.1361,
      "mean_token_accuracy": 0.9609280467033386,
      "num_tokens": 10927849.0,
      "step": 140
    },
    {
      "epoch": 0.22979397781299524,
      "grad_norm": 1.0677454471588135,
      "learning_rate": 3.55575963595269e-05,
      "loss": 0.1342,
      "mean_token_accuracy": 0.9626600503921509,
      "num_tokens": 11317280.0,
      "step": 145
    },
    {
      "epoch": 0.23771790808240886,
      "grad_norm": 0.7836272716522217,
      "learning_rate": 3.534955993548664e-05,
      "loss": 0.1156,
      "mean_token_accuracy": 0.9644288957118988,
      "num_tokens": 11708611.0,
      "step": 150
    },
    {
      "epoch": 0.24564183835182252,
      "grad_norm": 1.0319324731826782,
      "learning_rate": 3.513337150260695e-05,
      "loss": 0.1479,
      "mean_token_accuracy": 0.9606501817703247,
      "num_tokens": 12102608.0,
      "step": 155
    },
    {
      "epoch": 0.25356576862123614,
      "grad_norm": 0.8143829107284546,
      "learning_rate": 3.4909181726122425e-05,
      "loss": 0.1128,
      "mean_token_accuracy": 0.9646928489208222,
      "num_tokens": 12490263.0,
      "step": 160
    },
    {
      "epoch": 0.26148969889064977,
      "grad_norm": 0.7541149854660034,
      "learning_rate": 3.4677146847534146e-05,
      "loss": 0.0881,
      "mean_token_accuracy": 0.9717272222042084,
      "num_tokens": 12885149.0,
      "step": 165
    },
    {
      "epoch": 0.2694136291600634,
      "grad_norm": 0.5431804656982422,
      "learning_rate": 3.4437428575722406e-05,
      "loss": 0.0974,
      "mean_token_accuracy": 0.9715655505657196,
      "num_tokens": 13276483.0,
      "step": 170
    },
    {
      "epoch": 0.277337559429477,
      "grad_norm": 0.6050819158554077,
      "learning_rate": 3.419019397424923e-05,
      "loss": 0.0846,
      "mean_token_accuracy": 0.9722185015678406,
      "num_tokens": 13665007.0,
      "step": 175
    },
    {
      "epoch": 0.28526148969889065,
      "grad_norm": 0.7400597333908081,
      "learning_rate": 3.393561534492897e-05,
      "loss": 0.1094,
      "mean_token_accuracy": 0.9685170710086822,
      "num_tokens": 14056081.0,
      "step": 180
    },
    {
      "epoch": 0.2931854199683043,
      "grad_norm": 0.7055088877677917,
      "learning_rate": 3.3673870107748426e-05,
      "loss": 0.1283,
      "mean_token_accuracy": 0.9644022166728974,
      "num_tokens": 14441201.0,
      "step": 185
    },
    {
      "epoch": 0.3011093502377179,
      "grad_norm": 0.553695797920227,
      "learning_rate": 3.3405140677219984e-05,
      "loss": 0.1067,
      "mean_token_accuracy": 0.9692876458168029,
      "num_tokens": 14825085.0,
      "step": 190
    },
    {
      "epoch": 0.3090332805071315,
      "grad_norm": 0.6912402510643005,
      "learning_rate": 3.312961433525393e-05,
      "loss": 0.1071,
      "mean_token_accuracy": 0.9665566205978393,
      "num_tokens": 15214794.0,
      "step": 195
    },
    {
      "epoch": 0.31695721077654515,
      "grad_norm": 1.0452276468276978,
      "learning_rate": 3.284748310063868e-05,
      "loss": 0.1721,
      "mean_token_accuracy": 0.9547032237052917,
      "num_tokens": 15597986.0,
      "step": 200
    },
    {
      "epoch": 0.3248811410459588,
      "grad_norm": 0.5550786852836609,
      "learning_rate": 3.2558943595219744e-05,
      "loss": 0.1201,
      "mean_token_accuracy": 0.9660652875900269,
      "num_tokens": 15989335.0,
      "step": 205
    },
    {
      "epoch": 0.3328050713153724,
      "grad_norm": 0.44980260729789734,
      "learning_rate": 3.226419690687075e-05,
      "loss": 0.1102,
      "mean_token_accuracy": 0.9668643057346344,
      "num_tokens": 16385810.0,
      "step": 210
    },
    {
      "epoch": 0.34072900158478603,
      "grad_norm": 0.6570421457290649,
      "learning_rate": 3.196344844935208e-05,
      "loss": 0.1245,
      "mean_token_accuracy": 0.967197322845459,
      "num_tokens": 16777630.0,
      "step": 215
    },
    {
      "epoch": 0.3486529318541997,
      "grad_norm": 0.9588436484336853,
      "learning_rate": 3.165690781915463e-05,
      "loss": 0.1043,
      "mean_token_accuracy": 0.9675643861293792,
      "num_tokens": 17167482.0,
      "step": 220
    },
    {
      "epoch": 0.35657686212361334,
      "grad_norm": 0.8323602676391602,
      "learning_rate": 3.1344788649428664e-05,
      "loss": 0.1122,
      "mean_token_accuracy": 0.9644074141979218,
      "num_tokens": 17556201.0,
      "step": 225
    },
    {
      "epoch": 0.36450079239302696,
      "grad_norm": 1.5997897386550903,
      "learning_rate": 3.1027308461099314e-05,
      "loss": 0.1127,
      "mean_token_accuracy": 0.9671639263629913,
      "num_tokens": 17944746.0,
      "step": 230
    },
    {
      "epoch": 0.3724247226624406,
      "grad_norm": 0.46707582473754883,
      "learning_rate": 3.07046885112728e-05,
      "loss": 0.1384,
      "mean_token_accuracy": 0.96099693775177,
      "num_tokens": 18331814.0,
      "step": 235
    },
    {
      "epoch": 0.3803486529318542,
      "grad_norm": 0.561777651309967,
      "learning_rate": 3.037715363903869e-05,
      "loss": 0.1005,
      "mean_token_accuracy": 0.968315726518631,
      "num_tokens": 18722422.0,
      "step": 240
    },
    {
      "epoch": 0.38827258320126784,
      "grad_norm": 0.7177539467811584,
      "learning_rate": 3.004493210877593e-05,
      "loss": 0.1171,
      "mean_token_accuracy": 0.9664094805717468,
      "num_tokens": 19110658.0,
      "step": 245
    },
    {
      "epoch": 0.39619651347068147,
      "grad_norm": 0.8628878593444824,
      "learning_rate": 2.9708255451071633e-05,
      "loss": 0.1264,
      "mean_token_accuracy": 0.9644379198551178,
      "num_tokens": 19485097.0,
      "step": 250
    },
    {
      "epoch": 0.4041204437400951,
      "grad_norm": 0.9202964305877686,
      "learning_rate": 2.9367358301363695e-05,
      "loss": 0.0939,
      "mean_token_accuracy": 0.9741445899009704,
      "num_tokens": 19877219.0,
      "step": 255
    },
    {
      "epoch": 0.4120443740095087,
      "grad_norm": 1.1291356086730957,
      "learning_rate": 2.9022478236419525e-05,
      "loss": 0.0871,
      "mean_token_accuracy": 0.972924929857254,
      "num_tokens": 20271073.0,
      "step": 260
    },
    {
      "epoch": 0.41996830427892234,
      "grad_norm": 0.5406064987182617,
      "learning_rate": 2.867385560876492e-05,
      "loss": 0.1177,
      "mean_token_accuracy": 0.9652105033397674,
      "num_tokens": 20658930.0,
      "step": 265
    },
    {
      "epoch": 0.42789223454833597,
      "grad_norm": 0.5972470641136169,
      "learning_rate": 2.8321733379178497e-05,
      "loss": 0.1272,
      "mean_token_accuracy": 0.9624064087867736,
      "num_tokens": 21057127.0,
      "step": 270
    },
    {
      "epoch": 0.4358161648177496,
      "grad_norm": 0.7500821948051453,
      "learning_rate": 2.7966356947368384e-05,
      "loss": 0.1145,
      "mean_token_accuracy": 0.966503769159317,
      "num_tokens": 21439324.0,
      "step": 275
    },
    {
      "epoch": 0.4437400950871632,
      "grad_norm": 0.8380800485610962,
      "learning_rate": 2.7607973980949138e-05,
      "loss": 0.1634,
      "mean_token_accuracy": 0.9571461379528046,
      "num_tokens": 21827026.0,
      "step": 280
    },
    {
      "epoch": 0.45166402535657685,
      "grad_norm": 0.6657074689865112,
      "learning_rate": 2.7246834242838194e-05,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9761289060115814,
      "num_tokens": 22220416.0,
      "step": 285
    },
    {
      "epoch": 0.4595879556259905,
      "grad_norm": 1.1096389293670654,
      "learning_rate": 2.6883189417192004e-05,
      "loss": 0.1179,
      "mean_token_accuracy": 0.9632999897003174,
      "num_tokens": 22611802.0,
      "step": 290
    },
    {
      "epoch": 0.4675118858954041,
      "grad_norm": 0.6727749109268188,
      "learning_rate": 2.651729293400326e-05,
      "loss": 0.1056,
      "mean_token_accuracy": 0.9701432824134827,
      "num_tokens": 22998479.0,
      "step": 295
    },
    {
      "epoch": 0.4754358161648177,
      "grad_norm": 0.6313611268997192,
      "learning_rate": 2.6149399792481425e-05,
      "loss": 0.1166,
      "mean_token_accuracy": 0.9643904089927673,
      "num_tokens": 23384377.0,
      "step": 300
    },
    {
      "epoch": 0.48335974643423135,
      "grad_norm": 0.7332285642623901,
      "learning_rate": 2.57797663833396e-05,
      "loss": 0.0981,
      "mean_token_accuracy": 0.9713656902313232,
      "num_tokens": 23770163.0,
      "step": 305
    },
    {
      "epoch": 0.49128367670364503,
      "grad_norm": 0.6069464683532715,
      "learning_rate": 2.5408650310111714e-05,
      "loss": 0.1012,
      "mean_token_accuracy": 0.9690281748771667,
      "num_tokens": 24161677.0,
      "step": 310
    },
    {
      "epoch": 0.49920760697305866,
      "grad_norm": 0.5516354441642761,
      "learning_rate": 2.503631020962441e-05,
      "loss": 0.1029,
      "mean_token_accuracy": 0.9702194809913636,
      "num_tokens": 24545131.0,
      "step": 315
    },
    {
      "epoch": 0.5071315372424723,
      "grad_norm": 0.5970434546470642,
      "learning_rate": 2.4663005571748806e-05,
      "loss": 0.112,
      "mean_token_accuracy": 0.9674032151699066,
      "num_tokens": 24930689.0,
      "step": 320
    },
    {
      "epoch": 0.5150554675118859,
      "grad_norm": 0.9057714343070984,
      "learning_rate": 2.428899655855785e-05,
      "loss": 0.1253,
      "mean_token_accuracy": 0.9641950726509094,
      "num_tokens": 25317638.0,
      "step": 325
    },
    {
      "epoch": 0.5229793977812995,
      "grad_norm": 0.4484002888202667,
      "learning_rate": 2.3914543823015064e-05,
      "loss": 0.098,
      "mean_token_accuracy": 0.9691639423370362,
      "num_tokens": 25712776.0,
      "step": 330
    },
    {
      "epoch": 0.5309033280507132,
      "grad_norm": 0.6579416394233704,
      "learning_rate": 2.353990832732133e-05,
      "loss": 0.1223,
      "mean_token_accuracy": 0.9653512358665466,
      "num_tokens": 26097949.0,
      "step": 335
    },
    {
      "epoch": 0.5388272583201268,
      "grad_norm": 0.44268473982810974,
      "learning_rate": 2.3165351161046018e-05,
      "loss": 0.1041,
      "mean_token_accuracy": 0.9687828600406647,
      "num_tokens": 26496594.0,
      "step": 340
    },
    {
      "epoch": 0.5467511885895404,
      "grad_norm": 0.5640053153038025,
      "learning_rate": 2.279113335916947e-05,
      "loss": 0.1149,
      "mean_token_accuracy": 0.9681264221668243,
      "num_tokens": 26882411.0,
      "step": 345
    },
    {
      "epoch": 0.554675118858954,
      "grad_norm": 0.4891577363014221,
      "learning_rate": 2.2417515720163462e-05,
      "loss": 0.0866,
      "mean_token_accuracy": 0.9746865093708038,
      "num_tokens": 27265658.0,
      "step": 350
    },
    {
      "epoch": 0.5625990491283677,
      "grad_norm": 0.3668895959854126,
      "learning_rate": 2.20447586242365e-05,
      "loss": 0.0759,
      "mean_token_accuracy": 0.9763282477855683,
      "num_tokens": 27651242.0,
      "step": 355
    },
    {
      "epoch": 0.5705229793977813,
      "grad_norm": 0.9212456345558167,
      "learning_rate": 2.1673121851870622e-05,
      "loss": 0.1076,
      "mean_token_accuracy": 0.9682878911495209,
      "num_tokens": 28039286.0,
      "step": 360
    },
    {
      "epoch": 0.5784469096671949,
      "grad_norm": 0.5251045823097229,
      "learning_rate": 2.1302864402776154e-05,
      "loss": 0.0864,
      "mean_token_accuracy": 0.9751232445240021,
      "num_tokens": 28421022.0,
      "step": 365
    },
    {
      "epoch": 0.5863708399366085,
      "grad_norm": 0.7836456298828125,
      "learning_rate": 2.0934244315390614e-05,
      "loss": 0.0806,
      "mean_token_accuracy": 0.9762905418872834,
      "num_tokens": 28805146.0,
      "step": 370
    },
    {
      "epoch": 0.5942947702060222,
      "grad_norm": 0.8966671824455261,
      "learning_rate": 2.056751848704752e-05,
      "loss": 0.095,
      "mean_token_accuracy": 0.9731252431869507,
      "num_tokens": 29190736.0,
      "step": 375
    },
    {
      "epoch": 0.6022187004754358,
      "grad_norm": 0.5030308961868286,
      "learning_rate": 2.0202942494940434e-05,
      "loss": 0.1049,
      "mean_token_accuracy": 0.970298957824707,
      "num_tokens": 29575914.0,
      "step": 380
    },
    {
      "epoch": 0.6101426307448494,
      "grad_norm": 0.3350518047809601,
      "learning_rate": 1.984077041800711e-05,
      "loss": 0.1016,
      "mean_token_accuracy": 0.971675056219101,
      "num_tokens": 29962669.0,
      "step": 385
    },
    {
      "epoch": 0.618066561014263,
      "grad_norm": 0.825630247592926,
      "learning_rate": 1.9481254659857714e-05,
      "loss": 0.1067,
      "mean_token_accuracy": 0.9716610610485077,
      "num_tokens": 30351818.0,
      "step": 390
    },
    {
      "epoch": 0.6259904912836767,
      "grad_norm": 0.9551345705986023,
      "learning_rate": 1.912464577287066e-05,
      "loss": 0.128,
      "mean_token_accuracy": 0.9645381212234497,
      "num_tokens": 30735878.0,
      "step": 395
    },
    {
      "epoch": 0.6339144215530903,
      "grad_norm": 0.5807398557662964,
      "learning_rate": 1.8771192283578552e-05,
      "loss": 0.1058,
      "mean_token_accuracy": 0.970114243030548,
      "num_tokens": 31122981.0,
      "step": 400
    },
    {
      "epoch": 0.6418383518225039,
      "grad_norm": 0.5127037763595581,
      "learning_rate": 1.842114051946607e-05,
      "loss": 0.0854,
      "mean_token_accuracy": 0.9735840737819672,
      "num_tokens": 31506717.0,
      "step": 405
    },
    {
      "epoch": 0.6497622820919176,
      "grad_norm": 0.9288358688354492,
      "learning_rate": 1.8074734437300252e-05,
      "loss": 0.1189,
      "mean_token_accuracy": 0.967116779088974,
      "num_tokens": 31895528.0,
      "step": 410
    },
    {
      "epoch": 0.6576862123613312,
      "grad_norm": 0.818710446357727,
      "learning_rate": 1.773221545311313e-05,
      "loss": 0.1409,
      "mean_token_accuracy": 0.9593515217304229,
      "num_tokens": 32282510.0,
      "step": 415
    },
    {
      "epoch": 0.6656101426307448,
      "grad_norm": 0.6007797122001648,
      "learning_rate": 1.7393822273954927e-05,
      "loss": 0.0934,
      "mean_token_accuracy": 0.970532751083374,
      "num_tokens": 32666995.0,
      "step": 420
    },
    {
      "epoch": 0.6735340729001584,
      "grad_norm": 0.42770832777023315,
      "learning_rate": 1.7059790731535253e-05,
      "loss": 0.0935,
      "mean_token_accuracy": 0.9731517732143402,
      "num_tokens": 33053461.0,
      "step": 425
    },
    {
      "epoch": 0.6814580031695721,
      "grad_norm": 0.7949607372283936,
      "learning_rate": 1.6730353617868142e-05,
      "loss": 0.1154,
      "mean_token_accuracy": 0.9655673980712891,
      "num_tokens": 33441732.0,
      "step": 430
    },
    {
      "epoch": 0.6893819334389857,
      "grad_norm": 0.5710143446922302,
      "learning_rate": 1.6405740523035514e-05,
      "loss": 0.0973,
      "mean_token_accuracy": 0.9694910943508148,
      "num_tokens": 33829123.0,
      "step": 435
    },
    {
      "epoch": 0.6973058637083994,
      "grad_norm": 0.5229082703590393,
      "learning_rate": 1.6086177675182108e-05,
      "loss": 0.068,
      "mean_token_accuracy": 0.9778485476970673,
      "num_tokens": 34215039.0,
      "step": 440
    },
    {
      "epoch": 0.705229793977813,
      "grad_norm": 0.6268629431724548,
      "learning_rate": 1.5771887782853435e-05,
      "loss": 0.0884,
      "mean_token_accuracy": 0.9735526621341706,
      "num_tokens": 34607135.0,
      "step": 445
    },
    {
      "epoch": 0.7131537242472267,
      "grad_norm": 0.7254852652549744,
      "learning_rate": 1.5463089879786523e-05,
      "loss": 0.0962,
      "mean_token_accuracy": 0.9736709058284759,
      "num_tokens": 34987400.0,
      "step": 450
    },
    {
      "epoch": 0.7210776545166403,
      "grad_norm": 0.8859160542488098,
      "learning_rate": 1.5159999172261788e-05,
      "loss": 0.093,
      "mean_token_accuracy": 0.9735339641571045,
      "num_tokens": 35379373.0,
      "step": 455
    },
    {
      "epoch": 0.7290015847860539,
      "grad_norm": 0.7545599341392517,
      "learning_rate": 1.486282688912226e-05,
      "loss": 0.1203,
      "mean_token_accuracy": 0.9686921179294586,
      "num_tokens": 35774064.0,
      "step": 460
    },
    {
      "epoch": 0.7369255150554676,
      "grad_norm": 0.5699099898338318,
      "learning_rate": 1.457178013456476e-05,
      "loss": 0.0868,
      "mean_token_accuracy": 0.9763172626495361,
      "num_tokens": 36162675.0,
      "step": 465
    },
    {
      "epoch": 0.7448494453248812,
      "grad_norm": 0.984695315361023,
      "learning_rate": 1.4287061743805637e-05,
      "loss": 0.104,
      "mean_token_accuracy": 0.9697238504886627,
      "num_tokens": 36550874.0,
      "step": 470
    },
    {
      "epoch": 0.7527733755942948,
      "grad_norm": 0.6794852614402771,
      "learning_rate": 1.4008870141721574e-05,
      "loss": 0.1076,
      "mean_token_accuracy": 0.9720955193042755,
      "num_tokens": 36943782.0,
      "step": 475
    },
    {
      "epoch": 0.7606973058637084,
      "grad_norm": 0.5258546471595764,
      "learning_rate": 1.3737399204564117e-05,
      "loss": 0.096,
      "mean_token_accuracy": 0.9723799288272857,
      "num_tokens": 37327520.0,
      "step": 480
    },
    {
      "epoch": 0.768621236133122,
      "grad_norm": 0.5921935439109802,
      "learning_rate": 1.3472838124844096e-05,
      "loss": 0.0833,
      "mean_token_accuracy": 0.9756106197834015,
      "num_tokens": 37717199.0,
      "step": 485
    },
    {
      "epoch": 0.7765451664025357,
      "grad_norm": 0.5302931666374207,
      "learning_rate": 1.321537127948035e-05,
      "loss": 0.1059,
      "mean_token_accuracy": 0.9700763106346131,
      "num_tokens": 38108928.0,
      "step": 490
    },
    {
      "epoch": 0.7844690966719493,
      "grad_norm": 0.4625508487224579,
      "learning_rate": 1.2965178101304418e-05,
      "loss": 0.1001,
      "mean_token_accuracy": 0.9707379281520844,
      "num_tokens": 38498466.0,
      "step": 495
    },
    {
      "epoch": 0.7923930269413629,
      "grad_norm": 0.609122097492218,
      "learning_rate": 1.2722432954010879e-05,
      "loss": 0.095,
      "mean_token_accuracy": 0.9706707417964935,
      "num_tokens": 38886410.0,
      "step": 500
    },
    {
      "epoch": 0.8003169572107766,
      "grad_norm": 0.9062347412109375,
      "learning_rate": 1.2487305010640487e-05,
      "loss": 0.1062,
      "mean_token_accuracy": 0.9667559683322906,
      "num_tokens": 39278842.0,
      "step": 505
    },
    {
      "epoch": 0.8082408874801902,
      "grad_norm": 0.9213545322418213,
      "learning_rate": 1.2259958135680695e-05,
      "loss": 0.1205,
      "mean_token_accuracy": 0.9669659852981567,
      "num_tokens": 39666883.0,
      "step": 510
    },
    {
      "epoch": 0.8161648177496038,
      "grad_norm": 0.6698892712593079,
      "learning_rate": 1.2040550770865838e-05,
      "loss": 0.1043,
      "mean_token_accuracy": 0.9706094801425934,
      "num_tokens": 40056917.0,
      "step": 515
    },
    {
      "epoch": 0.8240887480190174,
      "grad_norm": 0.729739248752594,
      "learning_rate": 1.1829235824756539e-05,
      "loss": 0.1012,
      "mean_token_accuracy": 0.9677695989608764,
      "num_tokens": 40435895.0,
      "step": 520
    },
    {
      "epoch": 0.8320126782884311,
      "grad_norm": 0.5129750967025757,
      "learning_rate": 1.1626160566175199e-05,
      "loss": 0.0803,
      "mean_token_accuracy": 0.9758002817630768,
      "num_tokens": 40834814.0,
      "step": 525
    },
    {
      "epoch": 0.8399366085578447,
      "grad_norm": 0.6384918689727783,
      "learning_rate": 1.1431466521571993e-05,
      "loss": 0.1,
      "mean_token_accuracy": 0.9709450542926789,
      "num_tokens": 41211724.0,
      "step": 530
    },
    {
      "epoch": 0.8478605388272583,
      "grad_norm": 0.6116585731506348,
      "learning_rate": 1.1245289376392789e-05,
      "loss": 0.1128,
      "mean_token_accuracy": 0.9707423746585846,
      "num_tokens": 41605084.0,
      "step": 535
    },
    {
      "epoch": 0.8557844690966719,
      "grad_norm": 0.3281872570514679,
      "learning_rate": 1.1067758880517723e-05,
      "loss": 0.1101,
      "mean_token_accuracy": 0.9682676315307617,
      "num_tokens": 41993481.0,
      "step": 540
    },
    {
      "epoch": 0.8637083993660856,
      "grad_norm": 0.37608596682548523,
      "learning_rate": 1.0898998757836425e-05,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9784622848033905,
      "num_tokens": 42379094.0,
      "step": 545
    },
    {
      "epoch": 0.8716323296354992,
      "grad_norm": 0.42773088812828064,
      "learning_rate": 1.0739126620022823e-05,
      "loss": 0.0957,
      "mean_token_accuracy": 0.9720394492149353,
      "num_tokens": 42772765.0,
      "step": 550
    },
    {
      "epoch": 0.8795562599049128,
      "grad_norm": 0.7503993511199951,
      "learning_rate": 1.0588253884569649e-05,
      "loss": 0.0831,
      "mean_token_accuracy": 0.975706797838211,
      "num_tokens": 43164525.0,
      "step": 555
    },
    {
      "epoch": 0.8874801901743264,
      "grad_norm": 0.6069830060005188,
      "learning_rate": 1.0446485697139776e-05,
      "loss": 0.106,
      "mean_token_accuracy": 0.9685935854911805,
      "num_tokens": 43559276.0,
      "step": 560
    },
    {
      "epoch": 0.8954041204437401,
      "grad_norm": 0.6671760082244873,
      "learning_rate": 1.03139208582885e-05,
      "loss": 0.1067,
      "mean_token_accuracy": 0.9687129497528076,
      "num_tokens": 43947865.0,
      "step": 565
    },
    {
      "epoch": 0.9033280507131537,
      "grad_norm": 0.4149107336997986,
      "learning_rate": 1.0190651754607829e-05,
      "loss": 0.0771,
      "mean_token_accuracy": 0.9766448497772217,
      "num_tokens": 44336725.0,
      "step": 570
    },
    {
      "epoch": 0.9112519809825673,
      "grad_norm": 0.6251624822616577,
      "learning_rate": 1.0076764294340752e-05,
      "loss": 0.1169,
      "mean_token_accuracy": 0.9704745054244995,
      "num_tokens": 44727451.0,
      "step": 575
    },
    {
      "epoch": 0.919175911251981,
      "grad_norm": 1.3311954736709595,
      "learning_rate": 9.972337847510408e-06,
      "loss": 0.1047,
      "mean_token_accuracy": 0.9711760342121124,
      "num_tokens": 45123472.0,
      "step": 580
    },
    {
      "epoch": 0.9270998415213946,
      "grad_norm": 1.2829785346984863,
      "learning_rate": 9.877445190605806e-06,
      "loss": 0.144,
      "mean_token_accuracy": 0.9604130804538726,
      "num_tokens": 45516921.0,
      "step": 585
    },
    {
      "epoch": 0.9350237717908082,
      "grad_norm": 0.4952378571033478,
      "learning_rate": 9.792152455862712e-06,
      "loss": 0.086,
      "mean_token_accuracy": 0.9738980948925018,
      "num_tokens": 45903370.0,
      "step": 590
    },
    {
      "epoch": 0.9429477020602218,
      "grad_norm": 0.5334227085113525,
      "learning_rate": 9.716519085175029e-06,
      "loss": 0.0987,
      "mean_token_accuracy": 0.9714296877384185,
      "num_tokens": 46286779.0,
      "step": 595
    },
    {
      "epoch": 0.9508716323296355,
      "grad_norm": 0.32441264390945435,
      "learning_rate": 9.650597788668761e-06,
      "loss": 0.0822,
      "mean_token_accuracy": 0.9737054884433747,
      "num_tokens": 46676011.0,
      "step": 600
    },
    {
      "epoch": 0.9587955625990491,
      "grad_norm": 0.6578198075294495,
      "learning_rate": 9.59443450796747e-06,
      "loss": 0.0904,
      "mean_token_accuracy": 0.9721835911273956,
      "num_tokens": 47060484.0,
      "step": 605
    },
    {
      "epoch": 0.9667194928684627,
      "grad_norm": 0.5390242338180542,
      "learning_rate": 9.54806838417483e-06,
      "loss": 0.084,
      "mean_token_accuracy": 0.9759524583816528,
      "num_tokens": 47447497.0,
      "step": 610
    },
    {
      "epoch": 0.9746434231378764,
      "grad_norm": 0.6935057044029236,
      "learning_rate": 9.511531730596529e-06,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9739703357219696,
      "num_tokens": 47839755.0,
      "step": 615
    },
    {
      "epoch": 0.9825673534072901,
      "grad_norm": 0.8465091586112976,
      "learning_rate": 9.48485001022061e-06,
      "loss": 0.1073,
      "mean_token_accuracy": 0.9671664953231811,
      "num_tokens": 48224960.0,
      "step": 620
    },
    {
      "epoch": 0.9904912836767037,
      "grad_norm": 0.5479387640953064,
      "learning_rate": 9.468041817971918e-06,
      "loss": 0.0792,
      "mean_token_accuracy": 0.9757776200771332,
      "num_tokens": 48612318.0,
      "step": 625
    },
    {
      "epoch": 0.9984152139461173,
      "grad_norm": 0.6833792328834534,
      "learning_rate": 9.461118867752965e-06,
      "loss": 0.0771,
      "mean_token_accuracy": 0.9757878482341766,
      "num_tokens": 49003558.0,
      "step": 630
    }
  ],
  "logging_steps": 5,
  "max_steps": 630,
  "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": 3.1379649346593096e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}