{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 100,
  "global_step": 1250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.927482557296753,
      "epoch": 0.016,
      "grad_norm": 1.3331365585327148,
      "learning_rate": 4.736842105263158e-05,
      "loss": 2.9205,
      "mean_token_accuracy": 0.42432610020041467,
      "num_tokens": 21466.0,
      "step": 10
    },
    {
      "entropy": 2.7787948846817017,
      "epoch": 0.032,
      "grad_norm": 1.0274778604507446,
      "learning_rate": 0.0001,
      "loss": 2.7193,
      "mean_token_accuracy": 0.4424279421567917,
      "num_tokens": 44204.0,
      "step": 20
    },
    {
      "entropy": 2.7391002118587493,
      "epoch": 0.048,
      "grad_norm": 0.9975764751434326,
      "learning_rate": 0.00015263157894736845,
      "loss": 2.6235,
      "mean_token_accuracy": 0.46106590032577516,
      "num_tokens": 63982.0,
      "step": 30
    },
    {
      "entropy": 2.69979213476181,
      "epoch": 0.064,
      "grad_norm": 0.9850704669952393,
      "learning_rate": 0.00019999966405802826,
      "loss": 2.5825,
      "mean_token_accuracy": 0.4666088931262493,
      "num_tokens": 84531.0,
      "step": 40
    },
    {
      "entropy": 2.6038752436637878,
      "epoch": 0.08,
      "grad_norm": 0.9963482022285461,
      "learning_rate": 0.00019995935375248606,
      "loss": 2.5338,
      "mean_token_accuracy": 0.4776237912476063,
      "num_tokens": 104496.0,
      "step": 50
    },
    {
      "entropy": 2.6396988570690154,
      "epoch": 0.096,
      "grad_norm": 0.8940117955207825,
      "learning_rate": 0.000199851886084842,
      "loss": 2.5631,
      "mean_token_accuracy": 0.4724681057035923,
      "num_tokens": 125055.0,
      "step": 60
    },
    {
      "entropy": 2.624293786287308,
      "epoch": 0.112,
      "grad_norm": 0.7860824465751648,
      "learning_rate": 0.0001996773332568941,
      "loss": 2.5196,
      "mean_token_accuracy": 0.47606254518032076,
      "num_tokens": 145437.0,
      "step": 70
    },
    {
      "entropy": 2.523421919345856,
      "epoch": 0.128,
      "grad_norm": 0.7847617268562317,
      "learning_rate": 0.0001994358125413841,
      "loss": 2.438,
      "mean_token_accuracy": 0.4916156314313412,
      "num_tokens": 164918.0,
      "step": 80
    },
    {
      "entropy": 2.5297053515911103,
      "epoch": 0.144,
      "grad_norm": 0.8938744068145752,
      "learning_rate": 0.00019912748620320794,
      "loss": 2.477,
      "mean_token_accuracy": 0.4871461518108845,
      "num_tokens": 186602.0,
      "step": 90
    },
    {
      "entropy": 2.549694550037384,
      "epoch": 0.16,
      "grad_norm": 0.8062200546264648,
      "learning_rate": 0.00019875256139039902,
      "loss": 2.4538,
      "mean_token_accuracy": 0.48964215591549876,
      "num_tokens": 208334.0,
      "step": 100
    },
    {
      "epoch": 0.16,
      "eval_entropy": 2.5797194150777965,
      "eval_loss": 2.5428755283355713,
      "eval_mean_token_accuracy": 0.48654096172406125,
      "eval_num_tokens": 208334.0,
      "eval_runtime": 0.7687,
      "eval_samples_per_second": 130.09,
      "eval_steps_per_second": 16.912,
      "step": 100
    },
    {
      "entropy": 2.5369359493255614,
      "epoch": 0.176,
      "grad_norm": 0.7858350276947021,
      "learning_rate": 0.00019831128999495606,
      "loss": 2.451,
      "mean_token_accuracy": 0.49068020582199096,
      "num_tokens": 229573.0,
      "step": 110
    },
    {
      "entropy": 2.622505843639374,
      "epoch": 0.192,
      "grad_norm": 0.853503942489624,
      "learning_rate": 0.0001978039684836106,
      "loss": 2.513,
      "mean_token_accuracy": 0.4789179593324661,
      "num_tokens": 250739.0,
      "step": 120
    },
    {
      "entropy": 2.538061708211899,
      "epoch": 0.208,
      "grad_norm": 0.701422393321991,
      "learning_rate": 0.00019723093769864663,
      "loss": 2.4881,
      "mean_token_accuracy": 0.485482332110405,
      "num_tokens": 274310.0,
      "step": 130
    },
    {
      "entropy": 2.633000612258911,
      "epoch": 0.224,
      "grad_norm": 0.8620439171791077,
      "learning_rate": 0.00019659258262890683,
      "loss": 2.5312,
      "mean_token_accuracy": 0.4774567261338234,
      "num_tokens": 295034.0,
      "step": 140
    },
    {
      "entropy": 2.501206773519516,
      "epoch": 0.24,
      "grad_norm": 0.8027399778366089,
      "learning_rate": 0.00019588933215113926,
      "loss": 2.4184,
      "mean_token_accuracy": 0.4926304392516613,
      "num_tokens": 315487.0,
      "step": 150
    },
    {
      "entropy": 2.51069019138813,
      "epoch": 0.256,
      "grad_norm": 0.7904124855995178,
      "learning_rate": 0.00019512165874185767,
      "loss": 2.4492,
      "mean_token_accuracy": 0.49440020844340327,
      "num_tokens": 336958.0,
      "step": 160
    },
    {
      "entropy": 2.5472387611865996,
      "epoch": 0.272,
      "grad_norm": 0.8189327120780945,
      "learning_rate": 0.00019429007815990993,
      "loss": 2.4559,
      "mean_token_accuracy": 0.48381901159882545,
      "num_tokens": 358616.0,
      "step": 170
    },
    {
      "entropy": 2.547665923833847,
      "epoch": 0.288,
      "grad_norm": 0.7772969603538513,
      "learning_rate": 0.00019339514909996706,
      "loss": 2.4706,
      "mean_token_accuracy": 0.4824284970760345,
      "num_tokens": 379872.0,
      "step": 180
    },
    {
      "entropy": 2.504627192020416,
      "epoch": 0.304,
      "grad_norm": 0.6993403434753418,
      "learning_rate": 0.000192437472817166,
      "loss": 2.4443,
      "mean_token_accuracy": 0.4916023284196854,
      "num_tokens": 400768.0,
      "step": 190
    },
    {
      "entropy": 2.5959001183509827,
      "epoch": 0.32,
      "grad_norm": 0.8900535106658936,
      "learning_rate": 0.00019141769272315858,
      "loss": 2.5256,
      "mean_token_accuracy": 0.4826003320515156,
      "num_tokens": 420318.0,
      "step": 200
    },
    {
      "epoch": 0.32,
      "eval_entropy": 2.615577771113469,
      "eval_loss": 2.544278144836426,
      "eval_mean_token_accuracy": 0.4823318009193127,
      "eval_num_tokens": 420318.0,
      "eval_runtime": 0.6723,
      "eval_samples_per_second": 148.739,
      "eval_steps_per_second": 19.336,
      "step": 200
    },
    {
      "entropy": 2.5907178699970244,
      "epoch": 0.336,
      "grad_norm": 0.7852405309677124,
      "learning_rate": 0.00019033649395383702,
      "loss": 2.5025,
      "mean_token_accuracy": 0.48569325879216196,
      "num_tokens": 440493.0,
      "step": 210
    },
    {
      "entropy": 2.5373390048742293,
      "epoch": 0.352,
      "grad_norm": 0.8513323664665222,
      "learning_rate": 0.00018919460290902826,
      "loss": 2.4644,
      "mean_token_accuracy": 0.48737777322530745,
      "num_tokens": 460347.0,
      "step": 220
    },
    {
      "entropy": 2.5982253909111024,
      "epoch": 0.368,
      "grad_norm": 0.7951892614364624,
      "learning_rate": 0.00018799278676446423,
      "loss": 2.5088,
      "mean_token_accuracy": 0.4845408618450165,
      "num_tokens": 479686.0,
      "step": 230
    },
    {
      "entropy": 2.5408327341079713,
      "epoch": 0.384,
      "grad_norm": 0.818739652633667,
      "learning_rate": 0.0001867318529563574,
      "loss": 2.4453,
      "mean_token_accuracy": 0.49112383723258973,
      "num_tokens": 499382.0,
      "step": 240
    },
    {
      "entropy": 2.566703599691391,
      "epoch": 0.4,
      "grad_norm": 0.8289343118667603,
      "learning_rate": 0.00018541264863892754,
      "loss": 2.5047,
      "mean_token_accuracy": 0.48281258195638654,
      "num_tokens": 520865.0,
      "step": 250
    },
    {
      "entropy": 2.522556847333908,
      "epoch": 0.416,
      "grad_norm": 0.9309546947479248,
      "learning_rate": 0.000184036060115244,
      "loss": 2.4434,
      "mean_token_accuracy": 0.4828334227204323,
      "num_tokens": 540616.0,
      "step": 260
    },
    {
      "entropy": 2.5191095739603044,
      "epoch": 0.432,
      "grad_norm": 0.7696341276168823,
      "learning_rate": 0.00018260301224176558,
      "loss": 2.4318,
      "mean_token_accuracy": 0.4892562575638294,
      "num_tokens": 561098.0,
      "step": 270
    },
    {
      "entropy": 2.5569592356681823,
      "epoch": 0.448,
      "grad_norm": 0.7310771346092224,
      "learning_rate": 0.00018111446780697929,
      "loss": 2.505,
      "mean_token_accuracy": 0.476550355553627,
      "num_tokens": 581527.0,
      "step": 280
    },
    {
      "entropy": 2.6550574243068694,
      "epoch": 0.464,
      "grad_norm": 0.8539873957633972,
      "learning_rate": 0.00017957142688455362,
      "loss": 2.5309,
      "mean_token_accuracy": 0.479371577501297,
      "num_tokens": 601002.0,
      "step": 290
    },
    {
      "entropy": 2.5720873057842253,
      "epoch": 0.48,
      "grad_norm": 0.7261855006217957,
      "learning_rate": 0.00017797492616144256,
      "loss": 2.5289,
      "mean_token_accuracy": 0.4779372848570347,
      "num_tokens": 621836.0,
      "step": 300
    },
    {
      "epoch": 0.48,
      "eval_entropy": 2.58707464658297,
      "eval_loss": 2.5298879146575928,
      "eval_mean_token_accuracy": 0.4877227544784546,
      "eval_num_tokens": 621836.0,
      "eval_runtime": 0.6775,
      "eval_samples_per_second": 147.608,
      "eval_steps_per_second": 19.189,
      "step": 300
    },
    {
      "entropy": 2.56911381483078,
      "epoch": 0.496,
      "grad_norm": 0.8215358257293701,
      "learning_rate": 0.00017632603824139085,
      "loss": 2.4587,
      "mean_token_accuracy": 0.4831582002341747,
      "num_tokens": 642910.0,
      "step": 310
    },
    {
      "entropy": 2.518597739934921,
      "epoch": 0.512,
      "grad_norm": 0.7745339274406433,
      "learning_rate": 0.00017462587092430875,
      "loss": 2.4552,
      "mean_token_accuracy": 0.48742511570453645,
      "num_tokens": 665281.0,
      "step": 320
    },
    {
      "entropy": 2.5539629220962525,
      "epoch": 0.528,
      "grad_norm": 1.3013852834701538,
      "learning_rate": 0.00017287556646200018,
      "loss": 2.5049,
      "mean_token_accuracy": 0.48348471373319624,
      "num_tokens": 686496.0,
      "step": 330
    },
    {
      "entropy": 2.549721211194992,
      "epoch": 0.544,
      "grad_norm": 0.9040252566337585,
      "learning_rate": 0.00017107630079074478,
      "loss": 2.4749,
      "mean_token_accuracy": 0.49018326848745347,
      "num_tokens": 708041.0,
      "step": 340
    },
    {
      "entropy": 2.5657508850097654,
      "epoch": 0.56,
      "grad_norm": 0.7486541867256165,
      "learning_rate": 0.00016922928274124886,
      "loss": 2.496,
      "mean_token_accuracy": 0.4882775545120239,
      "num_tokens": 729373.0,
      "step": 350
    },
    {
      "entropy": 2.49460369348526,
      "epoch": 0.576,
      "grad_norm": 0.7519723176956177,
      "learning_rate": 0.00016733575322649657,
      "loss": 2.4358,
      "mean_token_accuracy": 0.4897538281977177,
      "num_tokens": 751428.0,
      "step": 360
    },
    {
      "entropy": 2.485343414545059,
      "epoch": 0.592,
      "grad_norm": 0.7465973496437073,
      "learning_rate": 0.00016539698440804661,
      "loss": 2.4446,
      "mean_token_accuracy": 0.4956702694296837,
      "num_tokens": 773898.0,
      "step": 370
    },
    {
      "entropy": 2.5664760768413544,
      "epoch": 0.608,
      "grad_norm": 0.792957603931427,
      "learning_rate": 0.0001634142788413346,
      "loss": 2.4411,
      "mean_token_accuracy": 0.49218829050660134,
      "num_tokens": 793921.0,
      "step": 380
    },
    {
      "entropy": 2.4981964468955993,
      "epoch": 0.624,
      "grad_norm": 0.9188303351402283,
      "learning_rate": 0.00016138896860055555,
      "loss": 2.4527,
      "mean_token_accuracy": 0.4891628086566925,
      "num_tokens": 814549.0,
      "step": 390
    },
    {
      "entropy": 2.5058595299720765,
      "epoch": 0.64,
      "grad_norm": 0.8261463642120361,
      "learning_rate": 0.0001593224143837142,
      "loss": 2.3825,
      "mean_token_accuracy": 0.49732586443424226,
      "num_tokens": 835738.0,
      "step": 400
    },
    {
      "epoch": 0.64,
      "eval_entropy": 2.5427963917072,
      "eval_loss": 2.528325080871582,
      "eval_mean_token_accuracy": 0.48722161238010114,
      "eval_num_tokens": 835738.0,
      "eval_runtime": 0.6794,
      "eval_samples_per_second": 147.184,
      "eval_steps_per_second": 19.134,
      "step": 400
    },
    {
      "entropy": 2.549647295475006,
      "epoch": 0.656,
      "grad_norm": 0.9669468402862549,
      "learning_rate": 0.00015721600459844468,
      "loss": 2.4992,
      "mean_token_accuracy": 0.4811323322355747,
      "num_tokens": 856308.0,
      "step": 410
    },
    {
      "entropy": 2.50358769595623,
      "epoch": 0.672,
      "grad_norm": 0.7736469507217407,
      "learning_rate": 0.0001550711544292131,
      "loss": 2.4233,
      "mean_token_accuracy": 0.4960208125412464,
      "num_tokens": 877984.0,
      "step": 420
    },
    {
      "entropy": 2.513987493515015,
      "epoch": 0.688,
      "grad_norm": 1.0335294008255005,
      "learning_rate": 0.00015288930488653094,
      "loss": 2.4349,
      "mean_token_accuracy": 0.490431859344244,
      "num_tokens": 899593.0,
      "step": 430
    },
    {
      "entropy": 2.5122825741767882,
      "epoch": 0.704,
      "grad_norm": 1.0326985120773315,
      "learning_rate": 0.00015067192183881658,
      "loss": 2.4435,
      "mean_token_accuracy": 0.49002100676298144,
      "num_tokens": 919549.0,
      "step": 440
    },
    {
      "entropy": 2.5058334708213805,
      "epoch": 0.72,
      "grad_norm": 0.8282021284103394,
      "learning_rate": 0.0001484204950275565,
      "loss": 2.3755,
      "mean_token_accuracy": 0.49881314411759375,
      "num_tokens": 938371.0,
      "step": 450
    },
    {
      "entropy": 2.5388923943042756,
      "epoch": 0.736,
      "grad_norm": 0.8473682403564453,
      "learning_rate": 0.0001461365370664276,
      "loss": 2.4695,
      "mean_token_accuracy": 0.48638317435979844,
      "num_tokens": 959081.0,
      "step": 460
    },
    {
      "entropy": 2.4723076283931733,
      "epoch": 0.752,
      "grad_norm": 0.964932918548584,
      "learning_rate": 0.00014382158242505234,
      "loss": 2.4151,
      "mean_token_accuracy": 0.4942440494894981,
      "num_tokens": 980709.0,
      "step": 470
    },
    {
      "entropy": 2.4699651658535005,
      "epoch": 0.768,
      "grad_norm": 0.8604968190193176,
      "learning_rate": 0.0001414771863980707,
      "loss": 2.4249,
      "mean_token_accuracy": 0.49896738603711127,
      "num_tokens": 998177.0,
      "step": 480
    },
    {
      "entropy": 2.5598580718040465,
      "epoch": 0.784,
      "grad_norm": 0.8044160604476929,
      "learning_rate": 0.00013910492406022033,
      "loss": 2.5119,
      "mean_token_accuracy": 0.48558894991874696,
      "num_tokens": 1019231.0,
      "step": 490
    },
    {
      "entropy": 2.5347262740135195,
      "epoch": 0.8,
      "grad_norm": 0.989012598991394,
      "learning_rate": 0.000136706389208128,
      "loss": 2.457,
      "mean_token_accuracy": 0.48754729256033896,
      "num_tokens": 1039963.0,
      "step": 500
    },
    {
      "epoch": 0.8,
      "eval_entropy": 2.5207965007195106,
      "eval_loss": 2.532066822052002,
      "eval_mean_token_accuracy": 0.4870225145266606,
      "eval_num_tokens": 1039963.0,
      "eval_runtime": 0.6776,
      "eval_samples_per_second": 147.573,
      "eval_steps_per_second": 19.185,
      "step": 500
    },
    {
      "entropy": 2.4576663255691527,
      "epoch": 0.816,
      "grad_norm": 0.8774695992469788,
      "learning_rate": 0.00013428319328952253,
      "loss": 2.3836,
      "mean_token_accuracy": 0.4995129629969597,
      "num_tokens": 1061370.0,
      "step": 510
    },
    {
      "entropy": 2.5250146806240084,
      "epoch": 0.832,
      "grad_norm": 0.7958005666732788,
      "learning_rate": 0.00013183696432058888,
      "loss": 2.4315,
      "mean_token_accuracy": 0.4982760787010193,
      "num_tokens": 1081842.0,
      "step": 520
    },
    {
      "entropy": 2.4409106403589247,
      "epoch": 0.848,
      "grad_norm": 0.7063387036323547,
      "learning_rate": 0.00012936934579219094,
      "loss": 2.3816,
      "mean_token_accuracy": 0.5006642855703831,
      "num_tokens": 1103737.0,
      "step": 530
    },
    {
      "entropy": 2.503674066066742,
      "epoch": 0.864,
      "grad_norm": 0.712315559387207,
      "learning_rate": 0.00012688199556569753,
      "loss": 2.4427,
      "mean_token_accuracy": 0.4878311850130558,
      "num_tokens": 1124676.0,
      "step": 540
    },
    {
      "entropy": 2.4937715351581575,
      "epoch": 0.88,
      "grad_norm": 0.7796215415000916,
      "learning_rate": 0.00012437658475915377,
      "loss": 2.4148,
      "mean_token_accuracy": 0.4976658411324024,
      "num_tokens": 1147148.0,
      "step": 550
    },
    {
      "entropy": 2.458594244718552,
      "epoch": 0.896,
      "grad_norm": 0.8076977133750916,
      "learning_rate": 0.00012185479662454595,
      "loss": 2.414,
      "mean_token_accuracy": 0.4989797443151474,
      "num_tokens": 1167948.0,
      "step": 560
    },
    {
      "entropy": 2.5129783391952514,
      "epoch": 0.912,
      "grad_norm": 0.9025003910064697,
      "learning_rate": 0.00011931832541691418,
      "loss": 2.4198,
      "mean_token_accuracy": 0.4928160272538662,
      "num_tokens": 1187896.0,
      "step": 570
    },
    {
      "entropy": 2.429825708270073,
      "epoch": 0.928,
      "grad_norm": 0.9303080439567566,
      "learning_rate": 0.00011676887525607271,
      "loss": 2.336,
      "mean_token_accuracy": 0.507265867292881,
      "num_tokens": 1207963.0,
      "step": 580
    },
    {
      "entropy": 2.483963069319725,
      "epoch": 0.944,
      "grad_norm": 0.9662640690803528,
      "learning_rate": 0.0001142081589817027,
      "loss": 2.4441,
      "mean_token_accuracy": 0.4951579734683037,
      "num_tokens": 1228312.0,
      "step": 590
    },
    {
      "entropy": 2.5128854513168335,
      "epoch": 0.96,
      "grad_norm": 0.7802348732948303,
      "learning_rate": 0.00011163789700258655,
      "loss": 2.481,
      "mean_token_accuracy": 0.4941797606647015,
      "num_tokens": 1250029.0,
      "step": 600
    },
    {
      "epoch": 0.96,
      "eval_entropy": 2.5629419730259824,
      "eval_loss": 2.526008367538452,
      "eval_mean_token_accuracy": 0.4862596323856941,
      "eval_num_tokens": 1250029.0,
      "eval_runtime": 0.6764,
      "eval_samples_per_second": 147.833,
      "eval_steps_per_second": 19.218,
      "step": 600
    },
    {
      "entropy": 2.45339612364769,
      "epoch": 0.976,
      "grad_norm": 0.7918068766593933,
      "learning_rate": 0.00010905981614075693,
      "loss": 2.3798,
      "mean_token_accuracy": 0.5059582263231277,
      "num_tokens": 1273572.0,
      "step": 610
    },
    {
      "entropy": 2.5436563432216643,
      "epoch": 0.992,
      "grad_norm": 0.9120656847953796,
      "learning_rate": 0.000106475648471337,
      "loss": 2.4771,
      "mean_token_accuracy": 0.49086978659033775,
      "num_tokens": 1295353.0,
      "step": 620
    },
    {
      "entropy": 2.410617971420288,
      "epoch": 1.008,
      "grad_norm": 0.7449939250946045,
      "learning_rate": 0.00010388713015885161,
      "loss": 2.2071,
      "mean_token_accuracy": 0.5270172961056232,
      "num_tokens": 1317590.0,
      "step": 630
    },
    {
      "entropy": 2.223151522874832,
      "epoch": 1.024,
      "grad_norm": 0.8070065975189209,
      "learning_rate": 0.00010129600029079072,
      "loss": 2.0441,
      "mean_token_accuracy": 0.5562543638050557,
      "num_tokens": 1338182.0,
      "step": 640
    },
    {
      "entropy": 2.1513119250535966,
      "epoch": 1.04,
      "grad_norm": 0.7470272779464722,
      "learning_rate": 9.870399970920932e-05,
      "loss": 2.0792,
      "mean_token_accuracy": 0.54966342151165,
      "num_tokens": 1358822.0,
      "step": 650
    },
    {
      "entropy": 2.2236103057861327,
      "epoch": 1.056,
      "grad_norm": 0.8093722462654114,
      "learning_rate": 9.611286984114841e-05,
      "loss": 2.0749,
      "mean_token_accuracy": 0.5545817121863366,
      "num_tokens": 1379005.0,
      "step": 660
    },
    {
      "entropy": 2.2356995731592177,
      "epoch": 1.072,
      "grad_norm": 0.8296709656715393,
      "learning_rate": 9.352435152866298e-05,
      "loss": 2.0653,
      "mean_token_accuracy": 0.5554739162325859,
      "num_tokens": 1401238.0,
      "step": 670
    },
    {
      "entropy": 2.240691155195236,
      "epoch": 1.088,
      "grad_norm": 0.864894688129425,
      "learning_rate": 9.09401838592431e-05,
      "loss": 2.1055,
      "mean_token_accuracy": 0.5520414739847184,
      "num_tokens": 1422418.0,
      "step": 680
    },
    {
      "entropy": 2.2082038104534147,
      "epoch": 1.104,
      "grad_norm": 1.003994345664978,
      "learning_rate": 8.836210299741346e-05,
      "loss": 2.0112,
      "mean_token_accuracy": 0.5623031251132489,
      "num_tokens": 1442486.0,
      "step": 690
    },
    {
      "entropy": 2.1362645506858824,
      "epoch": 1.12,
      "grad_norm": 0.9215587973594666,
      "learning_rate": 8.579184101829734e-05,
      "loss": 2.0474,
      "mean_token_accuracy": 0.5591939069330693,
      "num_tokens": 1465340.0,
      "step": 700
    },
    {
      "epoch": 1.12,
      "eval_entropy": 2.2982528393085184,
      "eval_loss": 2.5718441009521484,
      "eval_mean_token_accuracy": 0.4881514035738431,
      "eval_num_tokens": 1465340.0,
      "eval_runtime": 0.6785,
      "eval_samples_per_second": 147.375,
      "eval_steps_per_second": 19.159,
      "step": 700
    },
    {
      "entropy": 2.177516022324562,
      "epoch": 1.1360000000000001,
      "grad_norm": 0.8390331268310547,
      "learning_rate": 8.323112474392731e-05,
      "loss": 2.0602,
      "mean_token_accuracy": 0.5524092435836792,
      "num_tokens": 1484560.0,
      "step": 710
    },
    {
      "entropy": 2.2126996397972105,
      "epoch": 1.152,
      "grad_norm": 0.7812036275863647,
      "learning_rate": 8.068167458308582e-05,
      "loss": 2.08,
      "mean_token_accuracy": 0.5531082898378372,
      "num_tokens": 1506886.0,
      "step": 720
    },
    {
      "entropy": 2.172422558069229,
      "epoch": 1.168,
      "grad_norm": 0.9416456818580627,
      "learning_rate": 7.814520337545406e-05,
      "loss": 2.0252,
      "mean_token_accuracy": 0.5575959712266922,
      "num_tokens": 1529928.0,
      "step": 730
    },
    {
      "entropy": 2.161852565407753,
      "epoch": 1.184,
      "grad_norm": 1.1978275775909424,
      "learning_rate": 7.562341524084623e-05,
      "loss": 2.0087,
      "mean_token_accuracy": 0.5669111452996731,
      "num_tokens": 1547619.0,
      "step": 740
    },
    {
      "entropy": 2.0998224556446075,
      "epoch": 1.2,
      "grad_norm": 0.9615102410316467,
      "learning_rate": 7.311800443430251e-05,
      "loss": 1.9751,
      "mean_token_accuracy": 0.5734243541955948,
      "num_tokens": 1569169.0,
      "step": 750
    },
    {
      "entropy": 2.205637964606285,
      "epoch": 1.216,
      "grad_norm": 0.838168740272522,
      "learning_rate": 7.06306542078091e-05,
      "loss": 2.0775,
      "mean_token_accuracy": 0.5475761353969574,
      "num_tokens": 1591379.0,
      "step": 760
    },
    {
      "entropy": 2.220550259947777,
      "epoch": 1.232,
      "grad_norm": 1.063684344291687,
      "learning_rate": 6.816303567941112e-05,
      "loss": 2.0495,
      "mean_token_accuracy": 0.558470205962658,
      "num_tokens": 1611202.0,
      "step": 770
    },
    {
      "entropy": 2.1949700146913527,
      "epoch": 1.248,
      "grad_norm": 0.814487099647522,
      "learning_rate": 6.571680671047749e-05,
      "loss": 2.059,
      "mean_token_accuracy": 0.5488512426614761,
      "num_tokens": 1633430.0,
      "step": 780
    },
    {
      "entropy": 2.1767837584018705,
      "epoch": 1.264,
      "grad_norm": 0.8867488503456116,
      "learning_rate": 6.329361079187199e-05,
      "loss": 2.0492,
      "mean_token_accuracy": 0.5580153577029705,
      "num_tokens": 1654807.0,
      "step": 790
    },
    {
      "entropy": 2.15340778529644,
      "epoch": 1.28,
      "grad_norm": 0.8776085376739502,
      "learning_rate": 6.08950759397797e-05,
      "loss": 2.0017,
      "mean_token_accuracy": 0.5655137121677398,
      "num_tokens": 1676241.0,
      "step": 800
    },
    {
      "epoch": 1.28,
      "eval_entropy": 2.318764512355511,
      "eval_loss": 2.570997953414917,
      "eval_mean_token_accuracy": 0.4826100170612335,
      "eval_num_tokens": 1676241.0,
      "eval_runtime": 0.6801,
      "eval_samples_per_second": 147.041,
      "eval_steps_per_second": 19.115,
      "step": 800
    },
    {
      "entropy": 2.224538028240204,
      "epoch": 1.296,
      "grad_norm": 0.9554702043533325,
      "learning_rate": 5.8522813601929324e-05,
      "loss": 2.0907,
      "mean_token_accuracy": 0.5463516399264335,
      "num_tokens": 1696836.0,
      "step": 810
    },
    {
      "entropy": 2.1161206275224687,
      "epoch": 1.312,
      "grad_norm": 0.9823859333992004,
      "learning_rate": 5.617841757494762e-05,
      "loss": 1.954,
      "mean_token_accuracy": 0.5701966688036919,
      "num_tokens": 1717750.0,
      "step": 820
    },
    {
      "entropy": 2.187773588299751,
      "epoch": 1.328,
      "grad_norm": 0.9549944996833801,
      "learning_rate": 5.386346293357242e-05,
      "loss": 2.0556,
      "mean_token_accuracy": 0.5585292562842369,
      "num_tokens": 1738383.0,
      "step": 830
    },
    {
      "entropy": 2.142790347337723,
      "epoch": 1.3439999999999999,
      "grad_norm": 1.1180299520492554,
      "learning_rate": 5.15795049724435e-05,
      "loss": 1.9954,
      "mean_token_accuracy": 0.561890059709549,
      "num_tokens": 1757450.0,
      "step": 840
    },
    {
      "entropy": 2.137694215774536,
      "epoch": 1.3599999999999999,
      "grad_norm": 0.8447200059890747,
      "learning_rate": 4.9328078161183464e-05,
      "loss": 1.9993,
      "mean_token_accuracy": 0.5649342782795429,
      "num_tokens": 1777797.0,
      "step": 850
    },
    {
      "entropy": 2.1595336198806763,
      "epoch": 1.376,
      "grad_norm": 1.0689432621002197,
      "learning_rate": 4.7110695113469085e-05,
      "loss": 2.0498,
      "mean_token_accuracy": 0.5661192789673806,
      "num_tokens": 1799140.0,
      "step": 860
    },
    {
      "entropy": 2.1561016231775283,
      "epoch": 1.392,
      "grad_norm": 0.8344489336013794,
      "learning_rate": 4.492884557078688e-05,
      "loss": 2.0489,
      "mean_token_accuracy": 0.5576602071523666,
      "num_tokens": 1820412.0,
      "step": 870
    },
    {
      "entropy": 2.1310487359762194,
      "epoch": 1.408,
      "grad_norm": 0.8696361184120178,
      "learning_rate": 4.278399540155536e-05,
      "loss": 1.9619,
      "mean_token_accuracy": 0.5695676565170288,
      "num_tokens": 1841046.0,
      "step": 880
    },
    {
      "entropy": 2.093841528892517,
      "epoch": 1.424,
      "grad_norm": 0.9314078688621521,
      "learning_rate": 4.0677585616285774e-05,
      "loss": 1.9841,
      "mean_token_accuracy": 0.5746666699647903,
      "num_tokens": 1861480.0,
      "step": 890
    },
    {
      "entropy": 2.175295126438141,
      "epoch": 1.44,
      "grad_norm": 0.7976773977279663,
      "learning_rate": 3.861103139944449e-05,
      "loss": 2.0547,
      "mean_token_accuracy": 0.5568442553281784,
      "num_tokens": 1886031.0,
      "step": 900
    },
    {
      "epoch": 1.44,
      "eval_entropy": 2.285807416989253,
      "eval_loss": 2.586611032485962,
      "eval_mean_token_accuracy": 0.48370331755051243,
      "eval_num_tokens": 1886031.0,
      "eval_runtime": 0.6775,
      "eval_samples_per_second": 147.605,
      "eval_steps_per_second": 19.189,
      "step": 900
    },
    {
      "entropy": 2.1779676496982576,
      "epoch": 1.456,
      "grad_norm": 1.0191247463226318,
      "learning_rate": 3.658572115866541e-05,
      "loss": 2.0442,
      "mean_token_accuracy": 0.5562367714941502,
      "num_tokens": 1905543.0,
      "step": 910
    },
    {
      "entropy": 2.1235474199056625,
      "epoch": 1.472,
      "grad_norm": 0.811194896697998,
      "learning_rate": 3.4603015591953395e-05,
      "loss": 2.0,
      "mean_token_accuracy": 0.5647696286439896,
      "num_tokens": 1928338.0,
      "step": 920
    },
    {
      "entropy": 2.2050245642662047,
      "epoch": 1.488,
      "grad_norm": 1.0259207487106323,
      "learning_rate": 3.266424677350346e-05,
      "loss": 2.076,
      "mean_token_accuracy": 0.5486814387142658,
      "num_tokens": 1948809.0,
      "step": 930
    },
    {
      "entropy": 2.13138128221035,
      "epoch": 1.504,
      "grad_norm": 0.9353992342948914,
      "learning_rate": 3.077071725875116e-05,
      "loss": 1.9515,
      "mean_token_accuracy": 0.571763351559639,
      "num_tokens": 1969617.0,
      "step": 940
    },
    {
      "entropy": 2.1393859505653383,
      "epoch": 1.52,
      "grad_norm": 0.8784123659133911,
      "learning_rate": 2.8923699209255284e-05,
      "loss": 1.997,
      "mean_token_accuracy": 0.5623179793357849,
      "num_tokens": 1991588.0,
      "step": 950
    },
    {
      "entropy": 2.1316426753997804,
      "epoch": 1.536,
      "grad_norm": 0.9279586672782898,
      "learning_rate": 2.712443353799984e-05,
      "loss": 1.9962,
      "mean_token_accuracy": 0.5668175779283047,
      "num_tokens": 2013970.0,
      "step": 960
    },
    {
      "entropy": 2.1412221670150755,
      "epoch": 1.552,
      "grad_norm": 1.0614557266235352,
      "learning_rate": 2.5374129075691265e-05,
      "loss": 1.981,
      "mean_token_accuracy": 0.5675079435110092,
      "num_tokens": 2033870.0,
      "step": 970
    },
    {
      "entropy": 2.1519955545663834,
      "epoch": 1.568,
      "grad_norm": 1.0409959554672241,
      "learning_rate": 2.3673961758609152e-05,
      "loss": 2.0254,
      "mean_token_accuracy": 0.5568312801420688,
      "num_tokens": 2053742.0,
      "step": 980
    },
    {
      "entropy": 2.13378230035305,
      "epoch": 1.584,
      "grad_norm": 1.051721215248108,
      "learning_rate": 2.2025073838557454e-05,
      "loss": 1.9988,
      "mean_token_accuracy": 0.5596944905817509,
      "num_tokens": 2073771.0,
      "step": 990
    },
    {
      "entropy": 2.139891889691353,
      "epoch": 1.6,
      "grad_norm": 0.9190446138381958,
      "learning_rate": 2.0428573115446392e-05,
      "loss": 2.0098,
      "mean_token_accuracy": 0.5630866177380085,
      "num_tokens": 2093165.0,
      "step": 1000
    },
    {
      "epoch": 1.6,
      "eval_entropy": 2.29392801798307,
      "eval_loss": 2.5830881595611572,
      "eval_mean_token_accuracy": 0.48317546798632693,
      "eval_num_tokens": 2093165.0,
      "eval_runtime": 0.6748,
      "eval_samples_per_second": 148.201,
      "eval_steps_per_second": 19.266,
      "step": 1000
    },
    {
      "entropy": 2.152450978755951,
      "epoch": 1.616,
      "grad_norm": 0.9870165586471558,
      "learning_rate": 1.8885532193020704e-05,
      "loss": 2.0398,
      "mean_token_accuracy": 0.5565521039068699,
      "num_tokens": 2113836.0,
      "step": 1010
    },
    {
      "entropy": 2.1866767942905425,
      "epoch": 1.6320000000000001,
      "grad_norm": 0.9748448133468628,
      "learning_rate": 1.739698775823442e-05,
      "loss": 2.0618,
      "mean_token_accuracy": 0.55533082857728,
      "num_tokens": 2135597.0,
      "step": 1020
    },
    {
      "entropy": 2.1291246324777604,
      "epoch": 1.6480000000000001,
      "grad_norm": 0.9417283535003662,
      "learning_rate": 1.5963939884756042e-05,
      "loss": 2.0084,
      "mean_token_accuracy": 0.5552340872585774,
      "num_tokens": 2155871.0,
      "step": 1030
    },
    {
      "entropy": 2.1946566879749296,
      "epoch": 1.6640000000000001,
      "grad_norm": 0.8462656736373901,
      "learning_rate": 1.4587351361072454e-05,
      "loss": 2.0577,
      "mean_token_accuracy": 0.5526181682944298,
      "num_tokens": 2177686.0,
      "step": 1040
    },
    {
      "entropy": 2.1540873527526854,
      "epoch": 1.6800000000000002,
      "grad_norm": 0.8386402726173401,
      "learning_rate": 1.326814704364262e-05,
      "loss": 1.9869,
      "mean_token_accuracy": 0.5597876310348511,
      "num_tokens": 2197554.0,
      "step": 1050
    },
    {
      "entropy": 2.158621978759766,
      "epoch": 1.696,
      "grad_norm": 0.9277564287185669,
      "learning_rate": 1.2007213235535786e-05,
      "loss": 2.0627,
      "mean_token_accuracy": 0.5556139029562474,
      "num_tokens": 2216537.0,
      "step": 1060
    },
    {
      "entropy": 2.174393981695175,
      "epoch": 1.712,
      "grad_norm": 1.1009163856506348,
      "learning_rate": 1.0805397090971737e-05,
      "loss": 2.0276,
      "mean_token_accuracy": 0.562708406150341,
      "num_tokens": 2236200.0,
      "step": 1070
    },
    {
      "entropy": 2.1389470905065537,
      "epoch": 1.728,
      "grad_norm": 0.989471435546875,
      "learning_rate": 9.663506046162985e-06,
      "loss": 2.0102,
      "mean_token_accuracy": 0.5642252482473851,
      "num_tokens": 2258460.0,
      "step": 1080
    },
    {
      "entropy": 2.1603562027215957,
      "epoch": 1.744,
      "grad_norm": 1.0297486782073975,
      "learning_rate": 8.582307276841462e-06,
      "loss": 2.0182,
      "mean_token_accuracy": 0.5630522392690182,
      "num_tokens": 2278818.0,
      "step": 1090
    },
    {
      "entropy": 2.1662287533283235,
      "epoch": 1.76,
      "grad_norm": 0.8877721428871155,
      "learning_rate": 7.562527182833978e-06,
      "loss": 2.0168,
      "mean_token_accuracy": 0.563526850938797,
      "num_tokens": 2299692.0,
      "step": 1100
    },
    {
      "epoch": 1.76,
      "eval_entropy": 2.3022796649199266,
      "eval_loss": 2.5798184871673584,
      "eval_mean_token_accuracy": 0.4844009096805866,
      "eval_num_tokens": 2299692.0,
      "eval_runtime": 0.6747,
      "eval_samples_per_second": 148.214,
      "eval_steps_per_second": 19.268,
      "step": 1100
    },
    {
      "entropy": 2.186119019985199,
      "epoch": 1.776,
      "grad_norm": 1.1785519123077393,
      "learning_rate": 6.604850900032955e-06,
      "loss": 2.0263,
      "mean_token_accuracy": 0.5561132155358791,
      "num_tokens": 2320606.0,
      "step": 1110
    },
    {
      "entropy": 2.1382477432489395,
      "epoch": 1.792,
      "grad_norm": 0.9997260570526123,
      "learning_rate": 5.7099218400900716e-06,
      "loss": 2.0155,
      "mean_token_accuracy": 0.5585696510970592,
      "num_tokens": 2341589.0,
      "step": 1120
    },
    {
      "entropy": 2.1318988502025604,
      "epoch": 1.808,
      "grad_norm": 1.0082330703735352,
      "learning_rate": 4.87834125814235e-06,
      "loss": 2.013,
      "mean_token_accuracy": 0.562056128680706,
      "num_tokens": 2364065.0,
      "step": 1130
    },
    {
      "entropy": 2.1655119955539703,
      "epoch": 1.8239999999999998,
      "grad_norm": 0.8680001497268677,
      "learning_rate": 4.1106678488607495e-06,
      "loss": 2.0232,
      "mean_token_accuracy": 0.5640067130327224,
      "num_tokens": 2385547.0,
      "step": 1140
    },
    {
      "entropy": 2.133523789048195,
      "epoch": 1.8399999999999999,
      "grad_norm": 0.8609693050384521,
      "learning_rate": 3.40741737109318e-06,
      "loss": 1.971,
      "mean_token_accuracy": 0.5636379756033421,
      "num_tokens": 2406399.0,
      "step": 1150
    },
    {
      "entropy": 2.140376663208008,
      "epoch": 1.8559999999999999,
      "grad_norm": 0.917898952960968,
      "learning_rate": 2.7690623013533976e-06,
      "loss": 2.0089,
      "mean_token_accuracy": 0.5577343553304672,
      "num_tokens": 2425661.0,
      "step": 1160
    },
    {
      "entropy": 2.1588161110877992,
      "epoch": 1.8719999999999999,
      "grad_norm": 0.9883831739425659,
      "learning_rate": 2.1960315163894075e-06,
      "loss": 2.0176,
      "mean_token_accuracy": 0.5622549809515476,
      "num_tokens": 2445659.0,
      "step": 1170
    },
    {
      "entropy": 2.2101395189762116,
      "epoch": 1.888,
      "grad_norm": 0.7731896042823792,
      "learning_rate": 1.6887100050439587e-06,
      "loss": 2.0781,
      "mean_token_accuracy": 0.5513335958123207,
      "num_tokens": 2467209.0,
      "step": 1180
    },
    {
      "entropy": 2.124671292304993,
      "epoch": 1.904,
      "grad_norm": 0.8258634805679321,
      "learning_rate": 1.2474386096010039e-06,
      "loss": 1.986,
      "mean_token_accuracy": 0.5669601626694203,
      "num_tokens": 2489020.0,
      "step": 1190
    },
    {
      "entropy": 2.1453059643507,
      "epoch": 1.92,
      "grad_norm": 0.9933258295059204,
      "learning_rate": 8.725137967920738e-07,
      "loss": 2.0125,
      "mean_token_accuracy": 0.5601302161812782,
      "num_tokens": 2508813.0,
      "step": 1200
    },
    {
      "epoch": 1.92,
      "eval_entropy": 2.2944095684931827,
      "eval_loss": 2.5806496143341064,
      "eval_mean_token_accuracy": 0.48303504632069516,
      "eval_num_tokens": 2508813.0,
      "eval_runtime": 0.6714,
      "eval_samples_per_second": 148.939,
      "eval_steps_per_second": 19.362,
      "step": 1200
    },
    {
      "entropy": 2.151768389344215,
      "epoch": 1.936,
      "grad_norm": 0.8982884287834167,
      "learning_rate": 5.64187458615939e-07,
      "loss": 1.9924,
      "mean_token_accuracy": 0.5615834034979343,
      "num_tokens": 2528858.0,
      "step": 1210
    },
    {
      "entropy": 2.140746533870697,
      "epoch": 1.952,
      "grad_norm": 1.1360710859298706,
      "learning_rate": 3.2266674310589273e-07,
      "loss": 1.9953,
      "mean_token_accuracy": 0.5639697797596455,
      "num_tokens": 2548500.0,
      "step": 1220
    },
    {
      "entropy": 2.2176344990730286,
      "epoch": 1.968,
      "grad_norm": 1.0805954933166504,
      "learning_rate": 1.481139151579991e-07,
      "loss": 2.0792,
      "mean_token_accuracy": 0.5535091876983642,
      "num_tokens": 2568981.0,
      "step": 1230
    },
    {
      "entropy": 2.1372744172811506,
      "epoch": 1.984,
      "grad_norm": 0.936008870601654,
      "learning_rate": 4.064624751394242e-08,
      "loss": 2.0289,
      "mean_token_accuracy": 0.5637350581586361,
      "num_tokens": 2591580.0,
      "step": 1240
    },
    {
      "entropy": 2.156083407998085,
      "epoch": 2.0,
      "grad_norm": 0.864385724067688,
      "learning_rate": 3.3594197175190745e-10,
      "loss": 2.0083,
      "mean_token_accuracy": 0.5578774020075798,
      "num_tokens": 2612778.0,
      "step": 1250
    }
  ],
  "logging_steps": 10,
  "max_steps": 1250,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.51189440079872e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}