{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1167,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02570694087403599,
      "grad_norm": 5.865927219390869,
      "learning_rate": 5e-06,
      "loss": 0.9141,
      "step": 10
    },
    {
      "epoch": 0.05141388174807198,
      "grad_norm": 1.573315978050232,
      "learning_rate": 5e-06,
      "loss": 0.7782,
      "step": 20
    },
    {
      "epoch": 0.07712082262210797,
      "grad_norm": 1.0980241298675537,
      "learning_rate": 5e-06,
      "loss": 0.7291,
      "step": 30
    },
    {
      "epoch": 0.10282776349614396,
      "grad_norm": 0.8817281723022461,
      "learning_rate": 5e-06,
      "loss": 0.7015,
      "step": 40
    },
    {
      "epoch": 0.12853470437017994,
      "grad_norm": 1.167167067527771,
      "learning_rate": 5e-06,
      "loss": 0.6798,
      "step": 50
    },
    {
      "epoch": 0.15424164524421594,
      "grad_norm": 2.2831735610961914,
      "learning_rate": 5e-06,
      "loss": 0.6661,
      "step": 60
    },
    {
      "epoch": 0.17994858611825193,
      "grad_norm": 0.8653743863105774,
      "learning_rate": 5e-06,
      "loss": 0.6524,
      "step": 70
    },
    {
      "epoch": 0.20565552699228792,
      "grad_norm": 0.9545003175735474,
      "learning_rate": 5e-06,
      "loss": 0.6527,
      "step": 80
    },
    {
      "epoch": 0.23136246786632392,
      "grad_norm": 0.9084785580635071,
      "learning_rate": 5e-06,
      "loss": 0.6451,
      "step": 90
    },
    {
      "epoch": 0.2570694087403599,
      "grad_norm": 0.5454579591751099,
      "learning_rate": 5e-06,
      "loss": 0.6408,
      "step": 100
    },
    {
      "epoch": 0.2827763496143959,
      "grad_norm": 0.5495653748512268,
      "learning_rate": 5e-06,
      "loss": 0.6224,
      "step": 110
    },
    {
      "epoch": 0.30848329048843187,
      "grad_norm": 0.6504859328269958,
      "learning_rate": 5e-06,
      "loss": 0.6313,
      "step": 120
    },
    {
      "epoch": 0.3341902313624679,
      "grad_norm": 0.828336238861084,
      "learning_rate": 5e-06,
      "loss": 0.6324,
      "step": 130
    },
    {
      "epoch": 0.35989717223650386,
      "grad_norm": 0.5564887523651123,
      "learning_rate": 5e-06,
      "loss": 0.6231,
      "step": 140
    },
    {
      "epoch": 0.3856041131105398,
      "grad_norm": 0.7280133962631226,
      "learning_rate": 5e-06,
      "loss": 0.617,
      "step": 150
    },
    {
      "epoch": 0.41131105398457585,
      "grad_norm": 0.5732916593551636,
      "learning_rate": 5e-06,
      "loss": 0.6161,
      "step": 160
    },
    {
      "epoch": 0.4370179948586118,
      "grad_norm": 0.5130185484886169,
      "learning_rate": 5e-06,
      "loss": 0.616,
      "step": 170
    },
    {
      "epoch": 0.46272493573264784,
      "grad_norm": 1.034699559211731,
      "learning_rate": 5e-06,
      "loss": 0.6186,
      "step": 180
    },
    {
      "epoch": 0.4884318766066838,
      "grad_norm": 0.6324091553688049,
      "learning_rate": 5e-06,
      "loss": 0.6078,
      "step": 190
    },
    {
      "epoch": 0.5141388174807198,
      "grad_norm": 0.5066195726394653,
      "learning_rate": 5e-06,
      "loss": 0.6054,
      "step": 200
    },
    {
      "epoch": 0.5398457583547558,
      "grad_norm": 0.5803439021110535,
      "learning_rate": 5e-06,
      "loss": 0.6103,
      "step": 210
    },
    {
      "epoch": 0.5655526992287918,
      "grad_norm": 0.49449267983436584,
      "learning_rate": 5e-06,
      "loss": 0.6072,
      "step": 220
    },
    {
      "epoch": 0.5912596401028277,
      "grad_norm": 0.553966760635376,
      "learning_rate": 5e-06,
      "loss": 0.6012,
      "step": 230
    },
    {
      "epoch": 0.6169665809768637,
      "grad_norm": 0.5949456095695496,
      "learning_rate": 5e-06,
      "loss": 0.5979,
      "step": 240
    },
    {
      "epoch": 0.6426735218508998,
      "grad_norm": 0.5329740047454834,
      "learning_rate": 5e-06,
      "loss": 0.5985,
      "step": 250
    },
    {
      "epoch": 0.6683804627249358,
      "grad_norm": 0.6510268449783325,
      "learning_rate": 5e-06,
      "loss": 0.5938,
      "step": 260
    },
    {
      "epoch": 0.6940874035989717,
      "grad_norm": 0.4840281009674072,
      "learning_rate": 5e-06,
      "loss": 0.6043,
      "step": 270
    },
    {
      "epoch": 0.7197943444730077,
      "grad_norm": 0.9149703979492188,
      "learning_rate": 5e-06,
      "loss": 0.5969,
      "step": 280
    },
    {
      "epoch": 0.7455012853470437,
      "grad_norm": 0.6490294933319092,
      "learning_rate": 5e-06,
      "loss": 0.5942,
      "step": 290
    },
    {
      "epoch": 0.7712082262210797,
      "grad_norm": 0.6224178671836853,
      "learning_rate": 5e-06,
      "loss": 0.5936,
      "step": 300
    },
    {
      "epoch": 0.7969151670951157,
      "grad_norm": 0.5585442185401917,
      "learning_rate": 5e-06,
      "loss": 0.5903,
      "step": 310
    },
    {
      "epoch": 0.8226221079691517,
      "grad_norm": 0.5131608247756958,
      "learning_rate": 5e-06,
      "loss": 0.5864,
      "step": 320
    },
    {
      "epoch": 0.8483290488431876,
      "grad_norm": 0.7199530005455017,
      "learning_rate": 5e-06,
      "loss": 0.5891,
      "step": 330
    },
    {
      "epoch": 0.8740359897172236,
      "grad_norm": 0.5370771884918213,
      "learning_rate": 5e-06,
      "loss": 0.5793,
      "step": 340
    },
    {
      "epoch": 0.8997429305912596,
      "grad_norm": 0.7190908193588257,
      "learning_rate": 5e-06,
      "loss": 0.5908,
      "step": 350
    },
    {
      "epoch": 0.9254498714652957,
      "grad_norm": 0.5482605695724487,
      "learning_rate": 5e-06,
      "loss": 0.5825,
      "step": 360
    },
    {
      "epoch": 0.9511568123393316,
      "grad_norm": 0.5382151007652283,
      "learning_rate": 5e-06,
      "loss": 0.585,
      "step": 370
    },
    {
      "epoch": 0.9768637532133676,
      "grad_norm": 0.5443570017814636,
      "learning_rate": 5e-06,
      "loss": 0.5859,
      "step": 380
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.5778310894966125,
      "eval_runtime": 35.0525,
      "eval_samples_per_second": 299.038,
      "eval_steps_per_second": 1.17,
      "step": 389
    },
    {
      "epoch": 1.0025706940874035,
      "grad_norm": 0.8967987298965454,
      "learning_rate": 5e-06,
      "loss": 0.5767,
      "step": 390
    },
    {
      "epoch": 1.0282776349614395,
      "grad_norm": 0.5881332159042358,
      "learning_rate": 5e-06,
      "loss": 0.546,
      "step": 400
    },
    {
      "epoch": 1.0539845758354756,
      "grad_norm": 0.7526522278785706,
      "learning_rate": 5e-06,
      "loss": 0.5478,
      "step": 410
    },
    {
      "epoch": 1.0796915167095116,
      "grad_norm": 0.5316684246063232,
      "learning_rate": 5e-06,
      "loss": 0.5436,
      "step": 420
    },
    {
      "epoch": 1.1053984575835476,
      "grad_norm": 0.49085283279418945,
      "learning_rate": 5e-06,
      "loss": 0.5406,
      "step": 430
    },
    {
      "epoch": 1.1311053984575836,
      "grad_norm": 0.670981228351593,
      "learning_rate": 5e-06,
      "loss": 0.5446,
      "step": 440
    },
    {
      "epoch": 1.1568123393316196,
      "grad_norm": 0.5618010759353638,
      "learning_rate": 5e-06,
      "loss": 0.545,
      "step": 450
    },
    {
      "epoch": 1.1825192802056554,
      "grad_norm": 0.6097432971000671,
      "learning_rate": 5e-06,
      "loss": 0.5405,
      "step": 460
    },
    {
      "epoch": 1.2082262210796915,
      "grad_norm": 0.5328357815742493,
      "learning_rate": 5e-06,
      "loss": 0.5434,
      "step": 470
    },
    {
      "epoch": 1.2339331619537275,
      "grad_norm": 0.5351125597953796,
      "learning_rate": 5e-06,
      "loss": 0.5458,
      "step": 480
    },
    {
      "epoch": 1.2596401028277635,
      "grad_norm": 0.7780609130859375,
      "learning_rate": 5e-06,
      "loss": 0.5419,
      "step": 490
    },
    {
      "epoch": 1.2853470437017995,
      "grad_norm": 0.5531721115112305,
      "learning_rate": 5e-06,
      "loss": 0.5399,
      "step": 500
    },
    {
      "epoch": 1.3110539845758356,
      "grad_norm": 0.7177889943122864,
      "learning_rate": 5e-06,
      "loss": 0.5429,
      "step": 510
    },
    {
      "epoch": 1.3367609254498714,
      "grad_norm": 0.6722058653831482,
      "learning_rate": 5e-06,
      "loss": 0.5372,
      "step": 520
    },
    {
      "epoch": 1.3624678663239074,
      "grad_norm": 0.515562117099762,
      "learning_rate": 5e-06,
      "loss": 0.5335,
      "step": 530
    },
    {
      "epoch": 1.3881748071979434,
      "grad_norm": 0.6760542392730713,
      "learning_rate": 5e-06,
      "loss": 0.5361,
      "step": 540
    },
    {
      "epoch": 1.4138817480719794,
      "grad_norm": 0.7421192526817322,
      "learning_rate": 5e-06,
      "loss": 0.5399,
      "step": 550
    },
    {
      "epoch": 1.4395886889460154,
      "grad_norm": 0.5604066848754883,
      "learning_rate": 5e-06,
      "loss": 0.5426,
      "step": 560
    },
    {
      "epoch": 1.4652956298200515,
      "grad_norm": 0.5541772246360779,
      "learning_rate": 5e-06,
      "loss": 0.5392,
      "step": 570
    },
    {
      "epoch": 1.4910025706940875,
      "grad_norm": 0.5291936993598938,
      "learning_rate": 5e-06,
      "loss": 0.5355,
      "step": 580
    },
    {
      "epoch": 1.5167095115681235,
      "grad_norm": 0.5513972043991089,
      "learning_rate": 5e-06,
      "loss": 0.5388,
      "step": 590
    },
    {
      "epoch": 1.5424164524421595,
      "grad_norm": 0.5410069227218628,
      "learning_rate": 5e-06,
      "loss": 0.5351,
      "step": 600
    },
    {
      "epoch": 1.5681233933161953,
      "grad_norm": 0.5600557327270508,
      "learning_rate": 5e-06,
      "loss": 0.5368,
      "step": 610
    },
    {
      "epoch": 1.5938303341902313,
      "grad_norm": 0.5075570344924927,
      "learning_rate": 5e-06,
      "loss": 0.5274,
      "step": 620
    },
    {
      "epoch": 1.6195372750642674,
      "grad_norm": 0.5874601006507874,
      "learning_rate": 5e-06,
      "loss": 0.5335,
      "step": 630
    },
    {
      "epoch": 1.6452442159383034,
      "grad_norm": 0.5573471188545227,
      "learning_rate": 5e-06,
      "loss": 0.5282,
      "step": 640
    },
    {
      "epoch": 1.6709511568123392,
      "grad_norm": 0.7181922793388367,
      "learning_rate": 5e-06,
      "loss": 0.5357,
      "step": 650
    },
    {
      "epoch": 1.6966580976863752,
      "grad_norm": 0.5899803638458252,
      "learning_rate": 5e-06,
      "loss": 0.5281,
      "step": 660
    },
    {
      "epoch": 1.7223650385604112,
      "grad_norm": 0.5040240287780762,
      "learning_rate": 5e-06,
      "loss": 0.532,
      "step": 670
    },
    {
      "epoch": 1.7480719794344473,
      "grad_norm": 0.5562026500701904,
      "learning_rate": 5e-06,
      "loss": 0.5356,
      "step": 680
    },
    {
      "epoch": 1.7737789203084833,
      "grad_norm": 0.5617057681083679,
      "learning_rate": 5e-06,
      "loss": 0.5275,
      "step": 690
    },
    {
      "epoch": 1.7994858611825193,
      "grad_norm": 0.5864099860191345,
      "learning_rate": 5e-06,
      "loss": 0.5357,
      "step": 700
    },
    {
      "epoch": 1.8251928020565553,
      "grad_norm": 0.6038815975189209,
      "learning_rate": 5e-06,
      "loss": 0.5279,
      "step": 710
    },
    {
      "epoch": 1.8508997429305913,
      "grad_norm": 0.4948534667491913,
      "learning_rate": 5e-06,
      "loss": 0.5288,
      "step": 720
    },
    {
      "epoch": 1.8766066838046274,
      "grad_norm": 0.4712887406349182,
      "learning_rate": 5e-06,
      "loss": 0.5292,
      "step": 730
    },
    {
      "epoch": 1.9023136246786634,
      "grad_norm": 0.5233219265937805,
      "learning_rate": 5e-06,
      "loss": 0.5271,
      "step": 740
    },
    {
      "epoch": 1.9280205655526992,
      "grad_norm": 0.7113652229309082,
      "learning_rate": 5e-06,
      "loss": 0.5277,
      "step": 750
    },
    {
      "epoch": 1.9537275064267352,
      "grad_norm": 0.5814751982688904,
      "learning_rate": 5e-06,
      "loss": 0.5268,
      "step": 760
    },
    {
      "epoch": 1.9794344473007712,
      "grad_norm": 0.530799388885498,
      "learning_rate": 5e-06,
      "loss": 0.5297,
      "step": 770
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5520427227020264,
      "eval_runtime": 36.0966,
      "eval_samples_per_second": 290.388,
      "eval_steps_per_second": 1.136,
      "step": 778
    },
    {
      "epoch": 2.005141388174807,
      "grad_norm": 0.7985607981681824,
      "learning_rate": 5e-06,
      "loss": 0.5175,
      "step": 780
    },
    {
      "epoch": 2.030848329048843,
      "grad_norm": 0.6567503809928894,
      "learning_rate": 5e-06,
      "loss": 0.4906,
      "step": 790
    },
    {
      "epoch": 2.056555269922879,
      "grad_norm": 0.5952147841453552,
      "learning_rate": 5e-06,
      "loss": 0.4893,
      "step": 800
    },
    {
      "epoch": 2.082262210796915,
      "grad_norm": 0.6156149506568909,
      "learning_rate": 5e-06,
      "loss": 0.4989,
      "step": 810
    },
    {
      "epoch": 2.107969151670951,
      "grad_norm": 0.48527252674102783,
      "learning_rate": 5e-06,
      "loss": 0.4874,
      "step": 820
    },
    {
      "epoch": 2.133676092544987,
      "grad_norm": 0.546338677406311,
      "learning_rate": 5e-06,
      "loss": 0.4944,
      "step": 830
    },
    {
      "epoch": 2.159383033419023,
      "grad_norm": 0.5232033133506775,
      "learning_rate": 5e-06,
      "loss": 0.4871,
      "step": 840
    },
    {
      "epoch": 2.185089974293059,
      "grad_norm": 0.5629530549049377,
      "learning_rate": 5e-06,
      "loss": 0.4928,
      "step": 850
    },
    {
      "epoch": 2.210796915167095,
      "grad_norm": 0.6175139546394348,
      "learning_rate": 5e-06,
      "loss": 0.4909,
      "step": 860
    },
    {
      "epoch": 2.236503856041131,
      "grad_norm": 0.5951898097991943,
      "learning_rate": 5e-06,
      "loss": 0.489,
      "step": 870
    },
    {
      "epoch": 2.2622107969151672,
      "grad_norm": 0.57947838306427,
      "learning_rate": 5e-06,
      "loss": 0.4838,
      "step": 880
    },
    {
      "epoch": 2.2879177377892033,
      "grad_norm": 0.6488301753997803,
      "learning_rate": 5e-06,
      "loss": 0.4892,
      "step": 890
    },
    {
      "epoch": 2.3136246786632393,
      "grad_norm": 0.6758255362510681,
      "learning_rate": 5e-06,
      "loss": 0.4958,
      "step": 900
    },
    {
      "epoch": 2.3393316195372753,
      "grad_norm": 0.5344276428222656,
      "learning_rate": 5e-06,
      "loss": 0.4904,
      "step": 910
    },
    {
      "epoch": 2.365038560411311,
      "grad_norm": 0.560171365737915,
      "learning_rate": 5e-06,
      "loss": 0.4949,
      "step": 920
    },
    {
      "epoch": 2.390745501285347,
      "grad_norm": 0.6933562159538269,
      "learning_rate": 5e-06,
      "loss": 0.4941,
      "step": 930
    },
    {
      "epoch": 2.416452442159383,
      "grad_norm": 0.5910319089889526,
      "learning_rate": 5e-06,
      "loss": 0.4943,
      "step": 940
    },
    {
      "epoch": 2.442159383033419,
      "grad_norm": 0.5552846789360046,
      "learning_rate": 5e-06,
      "loss": 0.4924,
      "step": 950
    },
    {
      "epoch": 2.467866323907455,
      "grad_norm": 0.4899492859840393,
      "learning_rate": 5e-06,
      "loss": 0.486,
      "step": 960
    },
    {
      "epoch": 2.493573264781491,
      "grad_norm": 0.6860612630844116,
      "learning_rate": 5e-06,
      "loss": 0.4907,
      "step": 970
    },
    {
      "epoch": 2.519280205655527,
      "grad_norm": 0.5359469652175903,
      "learning_rate": 5e-06,
      "loss": 0.496,
      "step": 980
    },
    {
      "epoch": 2.544987146529563,
      "grad_norm": 0.49907293915748596,
      "learning_rate": 5e-06,
      "loss": 0.4927,
      "step": 990
    },
    {
      "epoch": 2.570694087403599,
      "grad_norm": 0.5387750864028931,
      "learning_rate": 5e-06,
      "loss": 0.4859,
      "step": 1000
    },
    {
      "epoch": 2.596401028277635,
      "grad_norm": 0.5668651461601257,
      "learning_rate": 5e-06,
      "loss": 0.4908,
      "step": 1010
    },
    {
      "epoch": 2.622107969151671,
      "grad_norm": 0.5228598117828369,
      "learning_rate": 5e-06,
      "loss": 0.4894,
      "step": 1020
    },
    {
      "epoch": 2.6478149100257067,
      "grad_norm": 0.5581383109092712,
      "learning_rate": 5e-06,
      "loss": 0.4913,
      "step": 1030
    },
    {
      "epoch": 2.6735218508997427,
      "grad_norm": 0.5759613513946533,
      "learning_rate": 5e-06,
      "loss": 0.4883,
      "step": 1040
    },
    {
      "epoch": 2.6992287917737787,
      "grad_norm": 0.6597785949707031,
      "learning_rate": 5e-06,
      "loss": 0.4852,
      "step": 1050
    },
    {
      "epoch": 2.7249357326478147,
      "grad_norm": 0.572317361831665,
      "learning_rate": 5e-06,
      "loss": 0.4875,
      "step": 1060
    },
    {
      "epoch": 2.7506426735218508,
      "grad_norm": 0.48137277364730835,
      "learning_rate": 5e-06,
      "loss": 0.4862,
      "step": 1070
    },
    {
      "epoch": 2.776349614395887,
      "grad_norm": 0.5613889694213867,
      "learning_rate": 5e-06,
      "loss": 0.4866,
      "step": 1080
    },
    {
      "epoch": 2.802056555269923,
      "grad_norm": 0.5657880902290344,
      "learning_rate": 5e-06,
      "loss": 0.4882,
      "step": 1090
    },
    {
      "epoch": 2.827763496143959,
      "grad_norm": 0.48773303627967834,
      "learning_rate": 5e-06,
      "loss": 0.4857,
      "step": 1100
    },
    {
      "epoch": 2.853470437017995,
      "grad_norm": 0.5101090669631958,
      "learning_rate": 5e-06,
      "loss": 0.4844,
      "step": 1110
    },
    {
      "epoch": 2.879177377892031,
      "grad_norm": 0.5357456207275391,
      "learning_rate": 5e-06,
      "loss": 0.4877,
      "step": 1120
    },
    {
      "epoch": 2.904884318766067,
      "grad_norm": 0.49460962414741516,
      "learning_rate": 5e-06,
      "loss": 0.4853,
      "step": 1130
    },
    {
      "epoch": 2.930591259640103,
      "grad_norm": 0.5633171796798706,
      "learning_rate": 5e-06,
      "loss": 0.4856,
      "step": 1140
    },
    {
      "epoch": 2.956298200514139,
      "grad_norm": 0.5715927481651306,
      "learning_rate": 5e-06,
      "loss": 0.4866,
      "step": 1150
    },
    {
      "epoch": 2.982005141388175,
      "grad_norm": 0.5925987958908081,
      "learning_rate": 5e-06,
      "loss": 0.4908,
      "step": 1160
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5453311800956726,
      "eval_runtime": 35.1038,
      "eval_samples_per_second": 298.6,
      "eval_steps_per_second": 1.168,
      "step": 1167
    },
    {
      "epoch": 3.0,
      "step": 1167,
      "total_flos": 5.5102128176360325e+19,
      "train_loss": 0.5515584751593321,
      "train_runtime": 8384.6233,
      "train_samples_per_second": 71.254,
      "train_steps_per_second": 0.139
    }
  ],
  "logging_steps": 10,
  "max_steps": 1167,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 5.5102128176360325e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}