{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 625,
  "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
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.76891183669248e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}