{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 5238,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0057273768613974796,
      "grad_norm": 1.1893006680700424,
      "learning_rate": 5e-06,
      "loss": 0.7164,
      "step": 10
    },
    {
      "epoch": 0.011454753722794959,
      "grad_norm": 1.04026748119074,
      "learning_rate": 5e-06,
      "loss": 0.6034,
      "step": 20
    },
    {
      "epoch": 0.01718213058419244,
      "grad_norm": 0.947476798322524,
      "learning_rate": 5e-06,
      "loss": 0.5621,
      "step": 30
    },
    {
      "epoch": 0.022909507445589918,
      "grad_norm": 0.8781018637437837,
      "learning_rate": 5e-06,
      "loss": 0.5341,
      "step": 40
    },
    {
      "epoch": 0.0286368843069874,
      "grad_norm": 0.9580875402301294,
      "learning_rate": 5e-06,
      "loss": 0.526,
      "step": 50
    },
    {
      "epoch": 0.03436426116838488,
      "grad_norm": 0.8781914311552192,
      "learning_rate": 5e-06,
      "loss": 0.5078,
      "step": 60
    },
    {
      "epoch": 0.04009163802978236,
      "grad_norm": 0.671566687192882,
      "learning_rate": 5e-06,
      "loss": 0.5056,
      "step": 70
    },
    {
      "epoch": 0.045819014891179836,
      "grad_norm": 0.6015476511437352,
      "learning_rate": 5e-06,
      "loss": 0.5,
      "step": 80
    },
    {
      "epoch": 0.05154639175257732,
      "grad_norm": 0.6233828429107251,
      "learning_rate": 5e-06,
      "loss": 0.5062,
      "step": 90
    },
    {
      "epoch": 0.0572737686139748,
      "grad_norm": 0.5502218388168,
      "learning_rate": 5e-06,
      "loss": 0.4995,
      "step": 100
    },
    {
      "epoch": 0.06300114547537228,
      "grad_norm": 0.5727700188508118,
      "learning_rate": 5e-06,
      "loss": 0.5018,
      "step": 110
    },
    {
      "epoch": 0.06872852233676977,
      "grad_norm": 0.5825399902463676,
      "learning_rate": 5e-06,
      "loss": 0.4897,
      "step": 120
    },
    {
      "epoch": 0.07445589919816724,
      "grad_norm": 0.5794095605103436,
      "learning_rate": 5e-06,
      "loss": 0.4921,
      "step": 130
    },
    {
      "epoch": 0.08018327605956473,
      "grad_norm": 0.5905275386374104,
      "learning_rate": 5e-06,
      "loss": 0.4872,
      "step": 140
    },
    {
      "epoch": 0.0859106529209622,
      "grad_norm": 0.5323846340490436,
      "learning_rate": 5e-06,
      "loss": 0.4771,
      "step": 150
    },
    {
      "epoch": 0.09163802978235967,
      "grad_norm": 0.5827507453646251,
      "learning_rate": 5e-06,
      "loss": 0.4923,
      "step": 160
    },
    {
      "epoch": 0.09736540664375716,
      "grad_norm": 0.6052222320099511,
      "learning_rate": 5e-06,
      "loss": 0.4889,
      "step": 170
    },
    {
      "epoch": 0.10309278350515463,
      "grad_norm": 0.5461900186435344,
      "learning_rate": 5e-06,
      "loss": 0.4906,
      "step": 180
    },
    {
      "epoch": 0.10882016036655212,
      "grad_norm": 0.5446312759280234,
      "learning_rate": 5e-06,
      "loss": 0.4792,
      "step": 190
    },
    {
      "epoch": 0.1145475372279496,
      "grad_norm": 0.5282278298362526,
      "learning_rate": 5e-06,
      "loss": 0.4749,
      "step": 200
    },
    {
      "epoch": 0.12027491408934708,
      "grad_norm": 0.5564949237418149,
      "learning_rate": 5e-06,
      "loss": 0.476,
      "step": 210
    },
    {
      "epoch": 0.12600229095074456,
      "grad_norm": 0.5511775080408522,
      "learning_rate": 5e-06,
      "loss": 0.4795,
      "step": 220
    },
    {
      "epoch": 0.13172966781214204,
      "grad_norm": 0.5328672926489202,
      "learning_rate": 5e-06,
      "loss": 0.4768,
      "step": 230
    },
    {
      "epoch": 0.13745704467353953,
      "grad_norm": 0.5278974935096795,
      "learning_rate": 5e-06,
      "loss": 0.471,
      "step": 240
    },
    {
      "epoch": 0.143184421534937,
      "grad_norm": 0.5354440170377812,
      "learning_rate": 5e-06,
      "loss": 0.4713,
      "step": 250
    },
    {
      "epoch": 0.14891179839633448,
      "grad_norm": 0.5438446163693844,
      "learning_rate": 5e-06,
      "loss": 0.4696,
      "step": 260
    },
    {
      "epoch": 0.15463917525773196,
      "grad_norm": 0.5496174748297591,
      "learning_rate": 5e-06,
      "loss": 0.4655,
      "step": 270
    },
    {
      "epoch": 0.16036655211912945,
      "grad_norm": 0.5926866056184342,
      "learning_rate": 5e-06,
      "loss": 0.4747,
      "step": 280
    },
    {
      "epoch": 0.1660939289805269,
      "grad_norm": 0.5141247582508096,
      "learning_rate": 5e-06,
      "loss": 0.4648,
      "step": 290
    },
    {
      "epoch": 0.1718213058419244,
      "grad_norm": 0.5536073278932672,
      "learning_rate": 5e-06,
      "loss": 0.4659,
      "step": 300
    },
    {
      "epoch": 0.1775486827033219,
      "grad_norm": 0.5234564239929281,
      "learning_rate": 5e-06,
      "loss": 0.4581,
      "step": 310
    },
    {
      "epoch": 0.18327605956471935,
      "grad_norm": 0.5219512867012481,
      "learning_rate": 5e-06,
      "loss": 0.4577,
      "step": 320
    },
    {
      "epoch": 0.18900343642611683,
      "grad_norm": 0.5474679242801981,
      "learning_rate": 5e-06,
      "loss": 0.463,
      "step": 330
    },
    {
      "epoch": 0.19473081328751432,
      "grad_norm": 0.516308993975195,
      "learning_rate": 5e-06,
      "loss": 0.4606,
      "step": 340
    },
    {
      "epoch": 0.2004581901489118,
      "grad_norm": 0.5401317271229032,
      "learning_rate": 5e-06,
      "loss": 0.4722,
      "step": 350
    },
    {
      "epoch": 0.20618556701030927,
      "grad_norm": 0.5234223381679171,
      "learning_rate": 5e-06,
      "loss": 0.461,
      "step": 360
    },
    {
      "epoch": 0.21191294387170675,
      "grad_norm": 0.5475015406952914,
      "learning_rate": 5e-06,
      "loss": 0.4527,
      "step": 370
    },
    {
      "epoch": 0.21764032073310424,
      "grad_norm": 0.5376680018928803,
      "learning_rate": 5e-06,
      "loss": 0.4618,
      "step": 380
    },
    {
      "epoch": 0.22336769759450173,
      "grad_norm": 0.5410529332865766,
      "learning_rate": 5e-06,
      "loss": 0.454,
      "step": 390
    },
    {
      "epoch": 0.2290950744558992,
      "grad_norm": 0.5652526379991017,
      "learning_rate": 5e-06,
      "loss": 0.4656,
      "step": 400
    },
    {
      "epoch": 0.23482245131729668,
      "grad_norm": 0.5555688465804705,
      "learning_rate": 5e-06,
      "loss": 0.4594,
      "step": 410
    },
    {
      "epoch": 0.24054982817869416,
      "grad_norm": 0.5701732999281934,
      "learning_rate": 5e-06,
      "loss": 0.4662,
      "step": 420
    },
    {
      "epoch": 0.24627720504009165,
      "grad_norm": 0.5512026939695381,
      "learning_rate": 5e-06,
      "loss": 0.4683,
      "step": 430
    },
    {
      "epoch": 0.2520045819014891,
      "grad_norm": 0.553814123316092,
      "learning_rate": 5e-06,
      "loss": 0.463,
      "step": 440
    },
    {
      "epoch": 0.25773195876288657,
      "grad_norm": 0.5200195587347526,
      "learning_rate": 5e-06,
      "loss": 0.4609,
      "step": 450
    },
    {
      "epoch": 0.2634593356242841,
      "grad_norm": 0.550688588791354,
      "learning_rate": 5e-06,
      "loss": 0.4457,
      "step": 460
    },
    {
      "epoch": 0.26918671248568155,
      "grad_norm": 0.5863844516267137,
      "learning_rate": 5e-06,
      "loss": 0.4632,
      "step": 470
    },
    {
      "epoch": 0.27491408934707906,
      "grad_norm": 0.5729329474663841,
      "learning_rate": 5e-06,
      "loss": 0.4434,
      "step": 480
    },
    {
      "epoch": 0.2806414662084765,
      "grad_norm": 0.5523269014403068,
      "learning_rate": 5e-06,
      "loss": 0.4433,
      "step": 490
    },
    {
      "epoch": 0.286368843069874,
      "grad_norm": 0.5116583829791455,
      "learning_rate": 5e-06,
      "loss": 0.4562,
      "step": 500
    },
    {
      "epoch": 0.2920962199312715,
      "grad_norm": 0.5787963201989073,
      "learning_rate": 5e-06,
      "loss": 0.4654,
      "step": 510
    },
    {
      "epoch": 0.29782359679266895,
      "grad_norm": 0.5447902232207782,
      "learning_rate": 5e-06,
      "loss": 0.4298,
      "step": 520
    },
    {
      "epoch": 0.3035509736540664,
      "grad_norm": 0.5423741210811989,
      "learning_rate": 5e-06,
      "loss": 0.463,
      "step": 530
    },
    {
      "epoch": 0.30927835051546393,
      "grad_norm": 0.5232608979728346,
      "learning_rate": 5e-06,
      "loss": 0.4534,
      "step": 540
    },
    {
      "epoch": 0.3150057273768614,
      "grad_norm": 0.5745290999609551,
      "learning_rate": 5e-06,
      "loss": 0.4514,
      "step": 550
    },
    {
      "epoch": 0.3207331042382589,
      "grad_norm": 0.5974645168072564,
      "learning_rate": 5e-06,
      "loss": 0.4606,
      "step": 560
    },
    {
      "epoch": 0.32646048109965636,
      "grad_norm": 0.5487734496746278,
      "learning_rate": 5e-06,
      "loss": 0.4607,
      "step": 570
    },
    {
      "epoch": 0.3321878579610538,
      "grad_norm": 0.5441562004543612,
      "learning_rate": 5e-06,
      "loss": 0.4507,
      "step": 580
    },
    {
      "epoch": 0.33791523482245134,
      "grad_norm": 0.5573672899202607,
      "learning_rate": 5e-06,
      "loss": 0.4441,
      "step": 590
    },
    {
      "epoch": 0.3436426116838488,
      "grad_norm": 0.5308370142895599,
      "learning_rate": 5e-06,
      "loss": 0.4518,
      "step": 600
    },
    {
      "epoch": 0.34936998854524626,
      "grad_norm": 0.5117603002146386,
      "learning_rate": 5e-06,
      "loss": 0.4583,
      "step": 610
    },
    {
      "epoch": 0.3550973654066438,
      "grad_norm": 0.5574954390885013,
      "learning_rate": 5e-06,
      "loss": 0.4603,
      "step": 620
    },
    {
      "epoch": 0.36082474226804123,
      "grad_norm": 0.5606001506682383,
      "learning_rate": 5e-06,
      "loss": 0.4514,
      "step": 630
    },
    {
      "epoch": 0.3665521191294387,
      "grad_norm": 0.5569042632837081,
      "learning_rate": 5e-06,
      "loss": 0.4537,
      "step": 640
    },
    {
      "epoch": 0.3722794959908362,
      "grad_norm": 0.5585555273085546,
      "learning_rate": 5e-06,
      "loss": 0.4626,
      "step": 650
    },
    {
      "epoch": 0.37800687285223367,
      "grad_norm": 0.5029368495012907,
      "learning_rate": 5e-06,
      "loss": 0.4438,
      "step": 660
    },
    {
      "epoch": 0.3837342497136312,
      "grad_norm": 0.5901403440941212,
      "learning_rate": 5e-06,
      "loss": 0.456,
      "step": 670
    },
    {
      "epoch": 0.38946162657502864,
      "grad_norm": 0.5195926126583429,
      "learning_rate": 5e-06,
      "loss": 0.4445,
      "step": 680
    },
    {
      "epoch": 0.3951890034364261,
      "grad_norm": 0.5225881930388002,
      "learning_rate": 5e-06,
      "loss": 0.4439,
      "step": 690
    },
    {
      "epoch": 0.4009163802978236,
      "grad_norm": 0.5279076911326633,
      "learning_rate": 5e-06,
      "loss": 0.4518,
      "step": 700
    },
    {
      "epoch": 0.4066437571592211,
      "grad_norm": 0.5178025129238295,
      "learning_rate": 5e-06,
      "loss": 0.4444,
      "step": 710
    },
    {
      "epoch": 0.41237113402061853,
      "grad_norm": 0.5391197058842394,
      "learning_rate": 5e-06,
      "loss": 0.4417,
      "step": 720
    },
    {
      "epoch": 0.41809851088201605,
      "grad_norm": 0.5693350216341284,
      "learning_rate": 5e-06,
      "loss": 0.4531,
      "step": 730
    },
    {
      "epoch": 0.4238258877434135,
      "grad_norm": 0.5272168668986746,
      "learning_rate": 5e-06,
      "loss": 0.4507,
      "step": 740
    },
    {
      "epoch": 0.42955326460481097,
      "grad_norm": 0.5601282100265634,
      "learning_rate": 5e-06,
      "loss": 0.4395,
      "step": 750
    },
    {
      "epoch": 0.4352806414662085,
      "grad_norm": 0.5283833355102217,
      "learning_rate": 5e-06,
      "loss": 0.4392,
      "step": 760
    },
    {
      "epoch": 0.44100801832760594,
      "grad_norm": 0.5348435308559183,
      "learning_rate": 5e-06,
      "loss": 0.4571,
      "step": 770
    },
    {
      "epoch": 0.44673539518900346,
      "grad_norm": 0.5431552168715813,
      "learning_rate": 5e-06,
      "loss": 0.4532,
      "step": 780
    },
    {
      "epoch": 0.4524627720504009,
      "grad_norm": 0.545120281271903,
      "learning_rate": 5e-06,
      "loss": 0.4534,
      "step": 790
    },
    {
      "epoch": 0.4581901489117984,
      "grad_norm": 0.5387987858480611,
      "learning_rate": 5e-06,
      "loss": 0.4536,
      "step": 800
    },
    {
      "epoch": 0.4639175257731959,
      "grad_norm": 0.5430847301042485,
      "learning_rate": 5e-06,
      "loss": 0.4469,
      "step": 810
    },
    {
      "epoch": 0.46964490263459335,
      "grad_norm": 0.5311236600330211,
      "learning_rate": 5e-06,
      "loss": 0.4428,
      "step": 820
    },
    {
      "epoch": 0.4753722794959908,
      "grad_norm": 0.5560451338696027,
      "learning_rate": 5e-06,
      "loss": 0.4502,
      "step": 830
    },
    {
      "epoch": 0.48109965635738833,
      "grad_norm": 0.5470174916746855,
      "learning_rate": 5e-06,
      "loss": 0.4482,
      "step": 840
    },
    {
      "epoch": 0.4868270332187858,
      "grad_norm": 0.563263919253685,
      "learning_rate": 5e-06,
      "loss": 0.4446,
      "step": 850
    },
    {
      "epoch": 0.4925544100801833,
      "grad_norm": 0.573113515782645,
      "learning_rate": 5e-06,
      "loss": 0.4508,
      "step": 860
    },
    {
      "epoch": 0.49828178694158076,
      "grad_norm": 0.5231438549151345,
      "learning_rate": 5e-06,
      "loss": 0.45,
      "step": 870
    },
    {
      "epoch": 0.5040091638029782,
      "grad_norm": 0.5176148074624103,
      "learning_rate": 5e-06,
      "loss": 0.4348,
      "step": 880
    },
    {
      "epoch": 0.5097365406643757,
      "grad_norm": 0.5309016268362335,
      "learning_rate": 5e-06,
      "loss": 0.443,
      "step": 890
    },
    {
      "epoch": 0.5154639175257731,
      "grad_norm": 0.5021579023128154,
      "learning_rate": 5e-06,
      "loss": 0.448,
      "step": 900
    },
    {
      "epoch": 0.5211912943871707,
      "grad_norm": 0.559779127609092,
      "learning_rate": 5e-06,
      "loss": 0.444,
      "step": 910
    },
    {
      "epoch": 0.5269186712485682,
      "grad_norm": 0.5789927290923392,
      "learning_rate": 5e-06,
      "loss": 0.4493,
      "step": 920
    },
    {
      "epoch": 0.5326460481099656,
      "grad_norm": 0.5690148664602116,
      "learning_rate": 5e-06,
      "loss": 0.4541,
      "step": 930
    },
    {
      "epoch": 0.5383734249713631,
      "grad_norm": 0.5018200538904565,
      "learning_rate": 5e-06,
      "loss": 0.4437,
      "step": 940
    },
    {
      "epoch": 0.5441008018327605,
      "grad_norm": 0.5626094146626068,
      "learning_rate": 5e-06,
      "loss": 0.438,
      "step": 950
    },
    {
      "epoch": 0.5498281786941581,
      "grad_norm": 0.5453326758465812,
      "learning_rate": 5e-06,
      "loss": 0.4382,
      "step": 960
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.5731487204099626,
      "learning_rate": 5e-06,
      "loss": 0.443,
      "step": 970
    },
    {
      "epoch": 0.561282932416953,
      "grad_norm": 0.5469415847599124,
      "learning_rate": 5e-06,
      "loss": 0.4541,
      "step": 980
    },
    {
      "epoch": 0.5670103092783505,
      "grad_norm": 0.5216017929376741,
      "learning_rate": 5e-06,
      "loss": 0.447,
      "step": 990
    },
    {
      "epoch": 0.572737686139748,
      "grad_norm": 0.5633524684965274,
      "learning_rate": 5e-06,
      "loss": 0.4371,
      "step": 1000
    },
    {
      "epoch": 0.5784650630011455,
      "grad_norm": 0.5293660880153032,
      "learning_rate": 5e-06,
      "loss": 0.4383,
      "step": 1010
    },
    {
      "epoch": 0.584192439862543,
      "grad_norm": 0.5399877214818077,
      "learning_rate": 5e-06,
      "loss": 0.456,
      "step": 1020
    },
    {
      "epoch": 0.5899198167239404,
      "grad_norm": 0.5249271107308078,
      "learning_rate": 5e-06,
      "loss": 0.459,
      "step": 1030
    },
    {
      "epoch": 0.5956471935853379,
      "grad_norm": 0.5341902718265518,
      "learning_rate": 5e-06,
      "loss": 0.4379,
      "step": 1040
    },
    {
      "epoch": 0.6013745704467354,
      "grad_norm": 0.5681490785135558,
      "learning_rate": 5e-06,
      "loss": 0.445,
      "step": 1050
    },
    {
      "epoch": 0.6071019473081328,
      "grad_norm": 0.5894093830458618,
      "learning_rate": 5e-06,
      "loss": 0.4353,
      "step": 1060
    },
    {
      "epoch": 0.6128293241695304,
      "grad_norm": 0.5221738673506199,
      "learning_rate": 5e-06,
      "loss": 0.4412,
      "step": 1070
    },
    {
      "epoch": 0.6185567010309279,
      "grad_norm": 0.5797474198144454,
      "learning_rate": 5e-06,
      "loss": 0.443,
      "step": 1080
    },
    {
      "epoch": 0.6242840778923253,
      "grad_norm": 0.5717734893270968,
      "learning_rate": 5e-06,
      "loss": 0.4435,
      "step": 1090
    },
    {
      "epoch": 0.6300114547537228,
      "grad_norm": 0.4981452424201392,
      "learning_rate": 5e-06,
      "loss": 0.4294,
      "step": 1100
    },
    {
      "epoch": 0.6357388316151202,
      "grad_norm": 0.5341189709831258,
      "learning_rate": 5e-06,
      "loss": 0.4443,
      "step": 1110
    },
    {
      "epoch": 0.6414662084765178,
      "grad_norm": 0.5337916570686927,
      "learning_rate": 5e-06,
      "loss": 0.4388,
      "step": 1120
    },
    {
      "epoch": 0.6471935853379153,
      "grad_norm": 0.5191012616320769,
      "learning_rate": 5e-06,
      "loss": 0.4278,
      "step": 1130
    },
    {
      "epoch": 0.6529209621993127,
      "grad_norm": 0.5532648229910325,
      "learning_rate": 5e-06,
      "loss": 0.4445,
      "step": 1140
    },
    {
      "epoch": 0.6586483390607102,
      "grad_norm": 0.5376131317705378,
      "learning_rate": 5e-06,
      "loss": 0.4443,
      "step": 1150
    },
    {
      "epoch": 0.6643757159221076,
      "grad_norm": 0.5163417459667795,
      "learning_rate": 5e-06,
      "loss": 0.449,
      "step": 1160
    },
    {
      "epoch": 0.6701030927835051,
      "grad_norm": 0.5750744468716225,
      "learning_rate": 5e-06,
      "loss": 0.4424,
      "step": 1170
    },
    {
      "epoch": 0.6758304696449027,
      "grad_norm": 0.5691426147058902,
      "learning_rate": 5e-06,
      "loss": 0.4409,
      "step": 1180
    },
    {
      "epoch": 0.6815578465063001,
      "grad_norm": 0.572169838625891,
      "learning_rate": 5e-06,
      "loss": 0.4408,
      "step": 1190
    },
    {
      "epoch": 0.6872852233676976,
      "grad_norm": 0.562085920947936,
      "learning_rate": 5e-06,
      "loss": 0.4539,
      "step": 1200
    },
    {
      "epoch": 0.693012600229095,
      "grad_norm": 0.549680800307771,
      "learning_rate": 5e-06,
      "loss": 0.4408,
      "step": 1210
    },
    {
      "epoch": 0.6987399770904925,
      "grad_norm": 0.5268961102121411,
      "learning_rate": 5e-06,
      "loss": 0.4318,
      "step": 1220
    },
    {
      "epoch": 0.7044673539518901,
      "grad_norm": 0.546084380380316,
      "learning_rate": 5e-06,
      "loss": 0.4274,
      "step": 1230
    },
    {
      "epoch": 0.7101947308132875,
      "grad_norm": 0.5789125996714688,
      "learning_rate": 5e-06,
      "loss": 0.426,
      "step": 1240
    },
    {
      "epoch": 0.715922107674685,
      "grad_norm": 0.5552812309484606,
      "learning_rate": 5e-06,
      "loss": 0.4433,
      "step": 1250
    },
    {
      "epoch": 0.7216494845360825,
      "grad_norm": 0.5490841525640856,
      "learning_rate": 5e-06,
      "loss": 0.4347,
      "step": 1260
    },
    {
      "epoch": 0.7273768613974799,
      "grad_norm": 0.5287976784408353,
      "learning_rate": 5e-06,
      "loss": 0.4404,
      "step": 1270
    },
    {
      "epoch": 0.7331042382588774,
      "grad_norm": 0.5303023254262377,
      "learning_rate": 5e-06,
      "loss": 0.4195,
      "step": 1280
    },
    {
      "epoch": 0.738831615120275,
      "grad_norm": 0.5615661953108583,
      "learning_rate": 5e-06,
      "loss": 0.4399,
      "step": 1290
    },
    {
      "epoch": 0.7445589919816724,
      "grad_norm": 0.5392244469382096,
      "learning_rate": 5e-06,
      "loss": 0.4416,
      "step": 1300
    },
    {
      "epoch": 0.7502863688430699,
      "grad_norm": 0.5631926544247716,
      "learning_rate": 5e-06,
      "loss": 0.4458,
      "step": 1310
    },
    {
      "epoch": 0.7560137457044673,
      "grad_norm": 0.5321981965200037,
      "learning_rate": 5e-06,
      "loss": 0.4425,
      "step": 1320
    },
    {
      "epoch": 0.7617411225658648,
      "grad_norm": 0.5472839772157344,
      "learning_rate": 5e-06,
      "loss": 0.438,
      "step": 1330
    },
    {
      "epoch": 0.7674684994272624,
      "grad_norm": 0.5148846097662794,
      "learning_rate": 5e-06,
      "loss": 0.441,
      "step": 1340
    },
    {
      "epoch": 0.7731958762886598,
      "grad_norm": 0.5814136125437906,
      "learning_rate": 5e-06,
      "loss": 0.4422,
      "step": 1350
    },
    {
      "epoch": 0.7789232531500573,
      "grad_norm": 0.4865290153302818,
      "learning_rate": 5e-06,
      "loss": 0.4195,
      "step": 1360
    },
    {
      "epoch": 0.7846506300114547,
      "grad_norm": 0.5729362641564091,
      "learning_rate": 5e-06,
      "loss": 0.4252,
      "step": 1370
    },
    {
      "epoch": 0.7903780068728522,
      "grad_norm": 0.513721699658904,
      "learning_rate": 5e-06,
      "loss": 0.428,
      "step": 1380
    },
    {
      "epoch": 0.7961053837342497,
      "grad_norm": 0.5404966717471676,
      "learning_rate": 5e-06,
      "loss": 0.4321,
      "step": 1390
    },
    {
      "epoch": 0.8018327605956472,
      "grad_norm": 0.5705888817719007,
      "learning_rate": 5e-06,
      "loss": 0.4292,
      "step": 1400
    },
    {
      "epoch": 0.8075601374570447,
      "grad_norm": 0.5433900974318504,
      "learning_rate": 5e-06,
      "loss": 0.436,
      "step": 1410
    },
    {
      "epoch": 0.8132875143184422,
      "grad_norm": 0.5233288735400002,
      "learning_rate": 5e-06,
      "loss": 0.4347,
      "step": 1420
    },
    {
      "epoch": 0.8190148911798396,
      "grad_norm": 0.5315754725482877,
      "learning_rate": 5e-06,
      "loss": 0.4343,
      "step": 1430
    },
    {
      "epoch": 0.8247422680412371,
      "grad_norm": 0.540898996728165,
      "learning_rate": 5e-06,
      "loss": 0.4385,
      "step": 1440
    },
    {
      "epoch": 0.8304696449026346,
      "grad_norm": 0.4918837151161211,
      "learning_rate": 5e-06,
      "loss": 0.4187,
      "step": 1450
    },
    {
      "epoch": 0.8361970217640321,
      "grad_norm": 0.5167559153442192,
      "learning_rate": 5e-06,
      "loss": 0.4354,
      "step": 1460
    },
    {
      "epoch": 0.8419243986254296,
      "grad_norm": 0.5388428750370872,
      "learning_rate": 5e-06,
      "loss": 0.442,
      "step": 1470
    },
    {
      "epoch": 0.847651775486827,
      "grad_norm": 0.5478670963710769,
      "learning_rate": 5e-06,
      "loss": 0.432,
      "step": 1480
    },
    {
      "epoch": 0.8533791523482245,
      "grad_norm": 0.5551238913425003,
      "learning_rate": 5e-06,
      "loss": 0.4349,
      "step": 1490
    },
    {
      "epoch": 0.8591065292096219,
      "grad_norm": 0.5096809451223056,
      "learning_rate": 5e-06,
      "loss": 0.4337,
      "step": 1500
    },
    {
      "epoch": 0.8648339060710195,
      "grad_norm": 0.5485551648513652,
      "learning_rate": 5e-06,
      "loss": 0.4304,
      "step": 1510
    },
    {
      "epoch": 0.870561282932417,
      "grad_norm": 0.5439471352377651,
      "learning_rate": 5e-06,
      "loss": 0.4291,
      "step": 1520
    },
    {
      "epoch": 0.8762886597938144,
      "grad_norm": 0.5541220552945509,
      "learning_rate": 5e-06,
      "loss": 0.435,
      "step": 1530
    },
    {
      "epoch": 0.8820160366552119,
      "grad_norm": 0.5549824280277013,
      "learning_rate": 5e-06,
      "loss": 0.4455,
      "step": 1540
    },
    {
      "epoch": 0.8877434135166093,
      "grad_norm": 0.5160198263139852,
      "learning_rate": 5e-06,
      "loss": 0.4246,
      "step": 1550
    },
    {
      "epoch": 0.8934707903780069,
      "grad_norm": 0.5789101227791154,
      "learning_rate": 5e-06,
      "loss": 0.4459,
      "step": 1560
    },
    {
      "epoch": 0.8991981672394044,
      "grad_norm": 0.5619060606532335,
      "learning_rate": 5e-06,
      "loss": 0.4227,
      "step": 1570
    },
    {
      "epoch": 0.9049255441008018,
      "grad_norm": 0.5131288466424669,
      "learning_rate": 5e-06,
      "loss": 0.425,
      "step": 1580
    },
    {
      "epoch": 0.9106529209621993,
      "grad_norm": 0.5378774111576103,
      "learning_rate": 5e-06,
      "loss": 0.4339,
      "step": 1590
    },
    {
      "epoch": 0.9163802978235968,
      "grad_norm": 0.5496777521069192,
      "learning_rate": 5e-06,
      "loss": 0.4351,
      "step": 1600
    },
    {
      "epoch": 0.9221076746849943,
      "grad_norm": 0.5090132737788183,
      "learning_rate": 5e-06,
      "loss": 0.4439,
      "step": 1610
    },
    {
      "epoch": 0.9278350515463918,
      "grad_norm": 0.5537237167714347,
      "learning_rate": 5e-06,
      "loss": 0.4315,
      "step": 1620
    },
    {
      "epoch": 0.9335624284077892,
      "grad_norm": 0.5083065954607388,
      "learning_rate": 5e-06,
      "loss": 0.4209,
      "step": 1630
    },
    {
      "epoch": 0.9392898052691867,
      "grad_norm": 0.5504634632628532,
      "learning_rate": 5e-06,
      "loss": 0.4255,
      "step": 1640
    },
    {
      "epoch": 0.9450171821305842,
      "grad_norm": 0.497466661980522,
      "learning_rate": 5e-06,
      "loss": 0.4359,
      "step": 1650
    },
    {
      "epoch": 0.9507445589919816,
      "grad_norm": 0.5728710662746097,
      "learning_rate": 5e-06,
      "loss": 0.4298,
      "step": 1660
    },
    {
      "epoch": 0.9564719358533792,
      "grad_norm": 0.5382940747387163,
      "learning_rate": 5e-06,
      "loss": 0.4321,
      "step": 1670
    },
    {
      "epoch": 0.9621993127147767,
      "grad_norm": 0.5602645110237611,
      "learning_rate": 5e-06,
      "loss": 0.4376,
      "step": 1680
    },
    {
      "epoch": 0.9679266895761741,
      "grad_norm": 0.500936047855792,
      "learning_rate": 5e-06,
      "loss": 0.4138,
      "step": 1690
    },
    {
      "epoch": 0.9736540664375716,
      "grad_norm": 0.5586135425680717,
      "learning_rate": 5e-06,
      "loss": 0.4348,
      "step": 1700
    },
    {
      "epoch": 0.979381443298969,
      "grad_norm": 0.5493075698232323,
      "learning_rate": 5e-06,
      "loss": 0.4235,
      "step": 1710
    },
    {
      "epoch": 0.9851088201603666,
      "grad_norm": 0.5618879613810053,
      "learning_rate": 5e-06,
      "loss": 0.4331,
      "step": 1720
    },
    {
      "epoch": 0.9908361970217641,
      "grad_norm": 0.5518636911221633,
      "learning_rate": 5e-06,
      "loss": 0.4324,
      "step": 1730
    },
    {
      "epoch": 0.9965635738831615,
      "grad_norm": 0.540041363098471,
      "learning_rate": 5e-06,
      "loss": 0.4292,
      "step": 1740
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.4308567941188812,
      "eval_runtime": 442.3894,
      "eval_samples_per_second": 26.587,
      "eval_steps_per_second": 0.416,
      "step": 1746
    },
    {
      "epoch": 1.002290950744559,
      "grad_norm": 0.5315755979624516,
      "learning_rate": 5e-06,
      "loss": 0.4118,
      "step": 1750
    },
    {
      "epoch": 1.0080183276059564,
      "grad_norm": 0.5480247806796007,
      "learning_rate": 5e-06,
      "loss": 0.3744,
      "step": 1760
    },
    {
      "epoch": 1.013745704467354,
      "grad_norm": 0.5149756078279293,
      "learning_rate": 5e-06,
      "loss": 0.378,
      "step": 1770
    },
    {
      "epoch": 1.0194730813287514,
      "grad_norm": 0.5171540890283103,
      "learning_rate": 5e-06,
      "loss": 0.3702,
      "step": 1780
    },
    {
      "epoch": 1.0252004581901488,
      "grad_norm": 0.5125282145044338,
      "learning_rate": 5e-06,
      "loss": 0.3711,
      "step": 1790
    },
    {
      "epoch": 1.0309278350515463,
      "grad_norm": 0.5207422414484574,
      "learning_rate": 5e-06,
      "loss": 0.3665,
      "step": 1800
    },
    {
      "epoch": 1.036655211912944,
      "grad_norm": 0.5786010896530626,
      "learning_rate": 5e-06,
      "loss": 0.3787,
      "step": 1810
    },
    {
      "epoch": 1.0423825887743414,
      "grad_norm": 0.5464322830590085,
      "learning_rate": 5e-06,
      "loss": 0.3701,
      "step": 1820
    },
    {
      "epoch": 1.0481099656357389,
      "grad_norm": 0.5205995452066101,
      "learning_rate": 5e-06,
      "loss": 0.3775,
      "step": 1830
    },
    {
      "epoch": 1.0538373424971363,
      "grad_norm": 0.519243616807458,
      "learning_rate": 5e-06,
      "loss": 0.3796,
      "step": 1840
    },
    {
      "epoch": 1.0595647193585338,
      "grad_norm": 0.516879079058521,
      "learning_rate": 5e-06,
      "loss": 0.3682,
      "step": 1850
    },
    {
      "epoch": 1.0652920962199313,
      "grad_norm": 0.5222769824122077,
      "learning_rate": 5e-06,
      "loss": 0.3727,
      "step": 1860
    },
    {
      "epoch": 1.0710194730813287,
      "grad_norm": 0.5445361815315836,
      "learning_rate": 5e-06,
      "loss": 0.3808,
      "step": 1870
    },
    {
      "epoch": 1.0767468499427262,
      "grad_norm": 0.5036623685893024,
      "learning_rate": 5e-06,
      "loss": 0.372,
      "step": 1880
    },
    {
      "epoch": 1.0824742268041236,
      "grad_norm": 0.5348337621107334,
      "learning_rate": 5e-06,
      "loss": 0.3732,
      "step": 1890
    },
    {
      "epoch": 1.088201603665521,
      "grad_norm": 0.5060065384189831,
      "learning_rate": 5e-06,
      "loss": 0.3846,
      "step": 1900
    },
    {
      "epoch": 1.0939289805269188,
      "grad_norm": 0.5079630313899987,
      "learning_rate": 5e-06,
      "loss": 0.3734,
      "step": 1910
    },
    {
      "epoch": 1.0996563573883162,
      "grad_norm": 0.5280819989664591,
      "learning_rate": 5e-06,
      "loss": 0.3795,
      "step": 1920
    },
    {
      "epoch": 1.1053837342497137,
      "grad_norm": 0.5170704766163384,
      "learning_rate": 5e-06,
      "loss": 0.3813,
      "step": 1930
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.490210122517552,
      "learning_rate": 5e-06,
      "loss": 0.3675,
      "step": 1940
    },
    {
      "epoch": 1.1168384879725086,
      "grad_norm": 0.5132872988639963,
      "learning_rate": 5e-06,
      "loss": 0.3725,
      "step": 1950
    },
    {
      "epoch": 1.122565864833906,
      "grad_norm": 0.5345634201153158,
      "learning_rate": 5e-06,
      "loss": 0.3717,
      "step": 1960
    },
    {
      "epoch": 1.1282932416953035,
      "grad_norm": 0.509614500177815,
      "learning_rate": 5e-06,
      "loss": 0.376,
      "step": 1970
    },
    {
      "epoch": 1.134020618556701,
      "grad_norm": 0.533789127642796,
      "learning_rate": 5e-06,
      "loss": 0.3813,
      "step": 1980
    },
    {
      "epoch": 1.1397479954180985,
      "grad_norm": 0.5430074485285918,
      "learning_rate": 5e-06,
      "loss": 0.3836,
      "step": 1990
    },
    {
      "epoch": 1.145475372279496,
      "grad_norm": 0.5313196763751518,
      "learning_rate": 5e-06,
      "loss": 0.3733,
      "step": 2000
    },
    {
      "epoch": 1.1512027491408934,
      "grad_norm": 0.5044508200904723,
      "learning_rate": 5e-06,
      "loss": 0.3783,
      "step": 2010
    },
    {
      "epoch": 1.1569301260022908,
      "grad_norm": 0.5686318202507047,
      "learning_rate": 5e-06,
      "loss": 0.3752,
      "step": 2020
    },
    {
      "epoch": 1.1626575028636885,
      "grad_norm": 0.5467468838281937,
      "learning_rate": 5e-06,
      "loss": 0.3762,
      "step": 2030
    },
    {
      "epoch": 1.168384879725086,
      "grad_norm": 0.5149539692191945,
      "learning_rate": 5e-06,
      "loss": 0.3743,
      "step": 2040
    },
    {
      "epoch": 1.1741122565864834,
      "grad_norm": 0.5312244954262079,
      "learning_rate": 5e-06,
      "loss": 0.3746,
      "step": 2050
    },
    {
      "epoch": 1.179839633447881,
      "grad_norm": 0.5219006331104148,
      "learning_rate": 5e-06,
      "loss": 0.3723,
      "step": 2060
    },
    {
      "epoch": 1.1855670103092784,
      "grad_norm": 0.538164681785682,
      "learning_rate": 5e-06,
      "loss": 0.375,
      "step": 2070
    },
    {
      "epoch": 1.1912943871706758,
      "grad_norm": 0.5113415155626444,
      "learning_rate": 5e-06,
      "loss": 0.3808,
      "step": 2080
    },
    {
      "epoch": 1.1970217640320733,
      "grad_norm": 0.5321490376681038,
      "learning_rate": 5e-06,
      "loss": 0.3699,
      "step": 2090
    },
    {
      "epoch": 1.2027491408934707,
      "grad_norm": 0.5024285063567219,
      "learning_rate": 5e-06,
      "loss": 0.37,
      "step": 2100
    },
    {
      "epoch": 1.2084765177548682,
      "grad_norm": 0.500252002027109,
      "learning_rate": 5e-06,
      "loss": 0.3782,
      "step": 2110
    },
    {
      "epoch": 1.2142038946162657,
      "grad_norm": 0.4924458421349769,
      "learning_rate": 5e-06,
      "loss": 0.3743,
      "step": 2120
    },
    {
      "epoch": 1.2199312714776633,
      "grad_norm": 0.4994293087338354,
      "learning_rate": 5e-06,
      "loss": 0.3758,
      "step": 2130
    },
    {
      "epoch": 1.2256586483390608,
      "grad_norm": 0.5610101096098137,
      "learning_rate": 5e-06,
      "loss": 0.3805,
      "step": 2140
    },
    {
      "epoch": 1.2313860252004583,
      "grad_norm": 0.537084984018735,
      "learning_rate": 5e-06,
      "loss": 0.3803,
      "step": 2150
    },
    {
      "epoch": 1.2371134020618557,
      "grad_norm": 0.5371742151025276,
      "learning_rate": 5e-06,
      "loss": 0.38,
      "step": 2160
    },
    {
      "epoch": 1.2428407789232532,
      "grad_norm": 0.5349028198444167,
      "learning_rate": 5e-06,
      "loss": 0.3808,
      "step": 2170
    },
    {
      "epoch": 1.2485681557846506,
      "grad_norm": 0.5168637085023853,
      "learning_rate": 5e-06,
      "loss": 0.3762,
      "step": 2180
    },
    {
      "epoch": 1.254295532646048,
      "grad_norm": 0.521040039917101,
      "learning_rate": 5e-06,
      "loss": 0.3724,
      "step": 2190
    },
    {
      "epoch": 1.2600229095074456,
      "grad_norm": 0.547586207928514,
      "learning_rate": 5e-06,
      "loss": 0.379,
      "step": 2200
    },
    {
      "epoch": 1.265750286368843,
      "grad_norm": 0.5217616276100534,
      "learning_rate": 5e-06,
      "loss": 0.3801,
      "step": 2210
    },
    {
      "epoch": 1.2714776632302405,
      "grad_norm": 0.5119693142628771,
      "learning_rate": 5e-06,
      "loss": 0.3717,
      "step": 2220
    },
    {
      "epoch": 1.277205040091638,
      "grad_norm": 0.49171715528514415,
      "learning_rate": 5e-06,
      "loss": 0.3733,
      "step": 2230
    },
    {
      "epoch": 1.2829324169530354,
      "grad_norm": 0.5583188472803077,
      "learning_rate": 5e-06,
      "loss": 0.3848,
      "step": 2240
    },
    {
      "epoch": 1.2886597938144329,
      "grad_norm": 0.5384351613915103,
      "learning_rate": 5e-06,
      "loss": 0.3708,
      "step": 2250
    },
    {
      "epoch": 1.2943871706758305,
      "grad_norm": 0.5283801778798911,
      "learning_rate": 5e-06,
      "loss": 0.3737,
      "step": 2260
    },
    {
      "epoch": 1.300114547537228,
      "grad_norm": 0.516459149172914,
      "learning_rate": 5e-06,
      "loss": 0.3778,
      "step": 2270
    },
    {
      "epoch": 1.3058419243986255,
      "grad_norm": 0.5064553031768843,
      "learning_rate": 5e-06,
      "loss": 0.3803,
      "step": 2280
    },
    {
      "epoch": 1.311569301260023,
      "grad_norm": 0.4992554472760309,
      "learning_rate": 5e-06,
      "loss": 0.3644,
      "step": 2290
    },
    {
      "epoch": 1.3172966781214204,
      "grad_norm": 0.5351102245997288,
      "learning_rate": 5e-06,
      "loss": 0.3754,
      "step": 2300
    },
    {
      "epoch": 1.3230240549828178,
      "grad_norm": 0.5744380836159825,
      "learning_rate": 5e-06,
      "loss": 0.3773,
      "step": 2310
    },
    {
      "epoch": 1.3287514318442153,
      "grad_norm": 0.5390997556454836,
      "learning_rate": 5e-06,
      "loss": 0.3702,
      "step": 2320
    },
    {
      "epoch": 1.3344788087056128,
      "grad_norm": 0.5159422766312324,
      "learning_rate": 5e-06,
      "loss": 0.3662,
      "step": 2330
    },
    {
      "epoch": 1.3402061855670104,
      "grad_norm": 0.5122893955215309,
      "learning_rate": 5e-06,
      "loss": 0.3662,
      "step": 2340
    },
    {
      "epoch": 1.345933562428408,
      "grad_norm": 0.5435516572685336,
      "learning_rate": 5e-06,
      "loss": 0.3692,
      "step": 2350
    },
    {
      "epoch": 1.3516609392898054,
      "grad_norm": 0.506519754552173,
      "learning_rate": 5e-06,
      "loss": 0.3775,
      "step": 2360
    },
    {
      "epoch": 1.3573883161512028,
      "grad_norm": 0.5382970585180483,
      "learning_rate": 5e-06,
      "loss": 0.3744,
      "step": 2370
    },
    {
      "epoch": 1.3631156930126003,
      "grad_norm": 0.5098385678364493,
      "learning_rate": 5e-06,
      "loss": 0.3817,
      "step": 2380
    },
    {
      "epoch": 1.3688430698739977,
      "grad_norm": 0.554626835720726,
      "learning_rate": 5e-06,
      "loss": 0.3846,
      "step": 2390
    },
    {
      "epoch": 1.3745704467353952,
      "grad_norm": 0.5148678581013812,
      "learning_rate": 5e-06,
      "loss": 0.3673,
      "step": 2400
    },
    {
      "epoch": 1.3802978235967927,
      "grad_norm": 0.5160690729715556,
      "learning_rate": 5e-06,
      "loss": 0.3733,
      "step": 2410
    },
    {
      "epoch": 1.38602520045819,
      "grad_norm": 0.49317947843655263,
      "learning_rate": 5e-06,
      "loss": 0.3676,
      "step": 2420
    },
    {
      "epoch": 1.3917525773195876,
      "grad_norm": 0.491680558889149,
      "learning_rate": 5e-06,
      "loss": 0.3726,
      "step": 2430
    },
    {
      "epoch": 1.397479954180985,
      "grad_norm": 0.5159563830253588,
      "learning_rate": 5e-06,
      "loss": 0.3722,
      "step": 2440
    },
    {
      "epoch": 1.4032073310423825,
      "grad_norm": 0.5153913293053918,
      "learning_rate": 5e-06,
      "loss": 0.378,
      "step": 2450
    },
    {
      "epoch": 1.40893470790378,
      "grad_norm": 0.5264135232150384,
      "learning_rate": 5e-06,
      "loss": 0.3788,
      "step": 2460
    },
    {
      "epoch": 1.4146620847651776,
      "grad_norm": 0.5382150315757761,
      "learning_rate": 5e-06,
      "loss": 0.3728,
      "step": 2470
    },
    {
      "epoch": 1.420389461626575,
      "grad_norm": 0.5310778600445981,
      "learning_rate": 5e-06,
      "loss": 0.3725,
      "step": 2480
    },
    {
      "epoch": 1.4261168384879725,
      "grad_norm": 0.5190694903204963,
      "learning_rate": 5e-06,
      "loss": 0.3708,
      "step": 2490
    },
    {
      "epoch": 1.43184421534937,
      "grad_norm": 0.5074266048391012,
      "learning_rate": 5e-06,
      "loss": 0.3816,
      "step": 2500
    },
    {
      "epoch": 1.4375715922107675,
      "grad_norm": 0.527668984455016,
      "learning_rate": 5e-06,
      "loss": 0.3692,
      "step": 2510
    },
    {
      "epoch": 1.443298969072165,
      "grad_norm": 0.5676081369199447,
      "learning_rate": 5e-06,
      "loss": 0.3726,
      "step": 2520
    },
    {
      "epoch": 1.4490263459335624,
      "grad_norm": 0.734177855082109,
      "learning_rate": 5e-06,
      "loss": 0.3626,
      "step": 2530
    },
    {
      "epoch": 1.4547537227949598,
      "grad_norm": 0.5303161196964384,
      "learning_rate": 5e-06,
      "loss": 0.3736,
      "step": 2540
    },
    {
      "epoch": 1.4604810996563573,
      "grad_norm": 0.4844944790072633,
      "learning_rate": 5e-06,
      "loss": 0.3717,
      "step": 2550
    },
    {
      "epoch": 1.466208476517755,
      "grad_norm": 0.5315356528221638,
      "learning_rate": 5e-06,
      "loss": 0.3739,
      "step": 2560
    },
    {
      "epoch": 1.4719358533791524,
      "grad_norm": 0.4772067027942158,
      "learning_rate": 5e-06,
      "loss": 0.3812,
      "step": 2570
    },
    {
      "epoch": 1.47766323024055,
      "grad_norm": 0.4939040642118836,
      "learning_rate": 5e-06,
      "loss": 0.3672,
      "step": 2580
    },
    {
      "epoch": 1.4833906071019474,
      "grad_norm": 0.508946651609973,
      "learning_rate": 5e-06,
      "loss": 0.3746,
      "step": 2590
    },
    {
      "epoch": 1.4891179839633448,
      "grad_norm": 0.5298078020553034,
      "learning_rate": 5e-06,
      "loss": 0.3733,
      "step": 2600
    },
    {
      "epoch": 1.4948453608247423,
      "grad_norm": 0.5432646094312185,
      "learning_rate": 5e-06,
      "loss": 0.3758,
      "step": 2610
    },
    {
      "epoch": 1.5005727376861397,
      "grad_norm": 0.5105782450661664,
      "learning_rate": 5e-06,
      "loss": 0.3805,
      "step": 2620
    },
    {
      "epoch": 1.5063001145475372,
      "grad_norm": 0.5382093019088371,
      "learning_rate": 5e-06,
      "loss": 0.381,
      "step": 2630
    },
    {
      "epoch": 1.5120274914089347,
      "grad_norm": 0.5204647981548546,
      "learning_rate": 5e-06,
      "loss": 0.3787,
      "step": 2640
    },
    {
      "epoch": 1.5177548682703321,
      "grad_norm": 0.5407064147138322,
      "learning_rate": 5e-06,
      "loss": 0.379,
      "step": 2650
    },
    {
      "epoch": 1.5234822451317296,
      "grad_norm": 0.5213399797673937,
      "learning_rate": 5e-06,
      "loss": 0.3815,
      "step": 2660
    },
    {
      "epoch": 1.529209621993127,
      "grad_norm": 0.5213816003555798,
      "learning_rate": 5e-06,
      "loss": 0.3881,
      "step": 2670
    },
    {
      "epoch": 1.5349369988545245,
      "grad_norm": 0.5142228705336312,
      "learning_rate": 5e-06,
      "loss": 0.3776,
      "step": 2680
    },
    {
      "epoch": 1.540664375715922,
      "grad_norm": 0.520697436811985,
      "learning_rate": 5e-06,
      "loss": 0.3806,
      "step": 2690
    },
    {
      "epoch": 1.5463917525773194,
      "grad_norm": 0.5637817682476384,
      "learning_rate": 5e-06,
      "loss": 0.3715,
      "step": 2700
    },
    {
      "epoch": 1.552119129438717,
      "grad_norm": 0.4915517576152862,
      "learning_rate": 5e-06,
      "loss": 0.3727,
      "step": 2710
    },
    {
      "epoch": 1.5578465063001146,
      "grad_norm": 0.5237728548563729,
      "learning_rate": 5e-06,
      "loss": 0.3709,
      "step": 2720
    },
    {
      "epoch": 1.563573883161512,
      "grad_norm": 0.7190217505740812,
      "learning_rate": 5e-06,
      "loss": 0.3803,
      "step": 2730
    },
    {
      "epoch": 1.5693012600229095,
      "grad_norm": 0.5209643317738942,
      "learning_rate": 5e-06,
      "loss": 0.371,
      "step": 2740
    },
    {
      "epoch": 1.575028636884307,
      "grad_norm": 0.5247668495585847,
      "learning_rate": 5e-06,
      "loss": 0.3808,
      "step": 2750
    },
    {
      "epoch": 1.5807560137457046,
      "grad_norm": 0.532703132334951,
      "learning_rate": 5e-06,
      "loss": 0.378,
      "step": 2760
    },
    {
      "epoch": 1.586483390607102,
      "grad_norm": 0.5117006044420915,
      "learning_rate": 5e-06,
      "loss": 0.3695,
      "step": 2770
    },
    {
      "epoch": 1.5922107674684995,
      "grad_norm": 0.5272216030753158,
      "learning_rate": 5e-06,
      "loss": 0.3666,
      "step": 2780
    },
    {
      "epoch": 1.597938144329897,
      "grad_norm": 0.5392842250351713,
      "learning_rate": 5e-06,
      "loss": 0.3783,
      "step": 2790
    },
    {
      "epoch": 1.6036655211912945,
      "grad_norm": 0.5211103103323186,
      "learning_rate": 5e-06,
      "loss": 0.3809,
      "step": 2800
    },
    {
      "epoch": 1.609392898052692,
      "grad_norm": 0.5434144298333962,
      "learning_rate": 5e-06,
      "loss": 0.3839,
      "step": 2810
    },
    {
      "epoch": 1.6151202749140894,
      "grad_norm": 0.6244778503333394,
      "learning_rate": 5e-06,
      "loss": 0.375,
      "step": 2820
    },
    {
      "epoch": 1.6208476517754868,
      "grad_norm": 0.7929144005943337,
      "learning_rate": 5e-06,
      "loss": 0.3843,
      "step": 2830
    },
    {
      "epoch": 1.6265750286368843,
      "grad_norm": 0.5076072639573785,
      "learning_rate": 5e-06,
      "loss": 0.3764,
      "step": 2840
    },
    {
      "epoch": 1.6323024054982818,
      "grad_norm": 0.499119817733024,
      "learning_rate": 5e-06,
      "loss": 0.3814,
      "step": 2850
    },
    {
      "epoch": 1.6380297823596792,
      "grad_norm": 0.5427237791683881,
      "learning_rate": 5e-06,
      "loss": 0.3732,
      "step": 2860
    },
    {
      "epoch": 1.6437571592210767,
      "grad_norm": 0.5072517280886356,
      "learning_rate": 5e-06,
      "loss": 0.3716,
      "step": 2870
    },
    {
      "epoch": 1.6494845360824741,
      "grad_norm": 0.5424242226752932,
      "learning_rate": 5e-06,
      "loss": 0.3725,
      "step": 2880
    },
    {
      "epoch": 1.6552119129438716,
      "grad_norm": 0.5239482084133175,
      "learning_rate": 5e-06,
      "loss": 0.3797,
      "step": 2890
    },
    {
      "epoch": 1.660939289805269,
      "grad_norm": 0.4877340871651499,
      "learning_rate": 5e-06,
      "loss": 0.3797,
      "step": 2900
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.5187696419666292,
      "learning_rate": 5e-06,
      "loss": 0.3732,
      "step": 2910
    },
    {
      "epoch": 1.6723940435280642,
      "grad_norm": 0.5165454581988194,
      "learning_rate": 5e-06,
      "loss": 0.3761,
      "step": 2920
    },
    {
      "epoch": 1.6781214203894617,
      "grad_norm": 0.5295455488567519,
      "learning_rate": 5e-06,
      "loss": 0.3732,
      "step": 2930
    },
    {
      "epoch": 1.6838487972508591,
      "grad_norm": 0.5356558582460169,
      "learning_rate": 5e-06,
      "loss": 0.3759,
      "step": 2940
    },
    {
      "epoch": 1.6895761741122566,
      "grad_norm": 0.5382234203233621,
      "learning_rate": 5e-06,
      "loss": 0.3809,
      "step": 2950
    },
    {
      "epoch": 1.695303550973654,
      "grad_norm": 0.5426839059978352,
      "learning_rate": 5e-06,
      "loss": 0.3818,
      "step": 2960
    },
    {
      "epoch": 1.7010309278350515,
      "grad_norm": 0.5589401799758412,
      "learning_rate": 5e-06,
      "loss": 0.3735,
      "step": 2970
    },
    {
      "epoch": 1.7067583046964492,
      "grad_norm": 0.5400070080828028,
      "learning_rate": 5e-06,
      "loss": 0.3715,
      "step": 2980
    },
    {
      "epoch": 1.7124856815578466,
      "grad_norm": 0.5125880859835712,
      "learning_rate": 5e-06,
      "loss": 0.3692,
      "step": 2990
    },
    {
      "epoch": 1.718213058419244,
      "grad_norm": 0.521750795912995,
      "learning_rate": 5e-06,
      "loss": 0.3734,
      "step": 3000
    },
    {
      "epoch": 1.7239404352806416,
      "grad_norm": 0.49977791549900996,
      "learning_rate": 5e-06,
      "loss": 0.3714,
      "step": 3010
    },
    {
      "epoch": 1.729667812142039,
      "grad_norm": 0.517176383552139,
      "learning_rate": 5e-06,
      "loss": 0.3798,
      "step": 3020
    },
    {
      "epoch": 1.7353951890034365,
      "grad_norm": 0.5569996765160103,
      "learning_rate": 5e-06,
      "loss": 0.3704,
      "step": 3030
    },
    {
      "epoch": 1.741122565864834,
      "grad_norm": 0.5152384355166261,
      "learning_rate": 5e-06,
      "loss": 0.3687,
      "step": 3040
    },
    {
      "epoch": 1.7468499427262314,
      "grad_norm": 0.5152215076607375,
      "learning_rate": 5e-06,
      "loss": 0.3711,
      "step": 3050
    },
    {
      "epoch": 1.7525773195876289,
      "grad_norm": 0.5209367759111013,
      "learning_rate": 5e-06,
      "loss": 0.367,
      "step": 3060
    },
    {
      "epoch": 1.7583046964490263,
      "grad_norm": 0.72478206720126,
      "learning_rate": 5e-06,
      "loss": 0.3753,
      "step": 3070
    },
    {
      "epoch": 1.7640320733104238,
      "grad_norm": 0.5108998808880466,
      "learning_rate": 5e-06,
      "loss": 0.3776,
      "step": 3080
    },
    {
      "epoch": 1.7697594501718212,
      "grad_norm": 0.49545669464589914,
      "learning_rate": 5e-06,
      "loss": 0.3691,
      "step": 3090
    },
    {
      "epoch": 1.7754868270332187,
      "grad_norm": 0.5370814607474198,
      "learning_rate": 5e-06,
      "loss": 0.3746,
      "step": 3100
    },
    {
      "epoch": 1.7812142038946162,
      "grad_norm": 0.6072170324437888,
      "learning_rate": 5e-06,
      "loss": 0.3729,
      "step": 3110
    },
    {
      "epoch": 1.7869415807560136,
      "grad_norm": 0.5271754924806036,
      "learning_rate": 5e-06,
      "loss": 0.3588,
      "step": 3120
    },
    {
      "epoch": 1.792668957617411,
      "grad_norm": 0.5754508589270141,
      "learning_rate": 5e-06,
      "loss": 0.3776,
      "step": 3130
    },
    {
      "epoch": 1.7983963344788088,
      "grad_norm": 0.5403755251129506,
      "learning_rate": 5e-06,
      "loss": 0.3762,
      "step": 3140
    },
    {
      "epoch": 1.8041237113402062,
      "grad_norm": 0.5138981496909732,
      "learning_rate": 5e-06,
      "loss": 0.3711,
      "step": 3150
    },
    {
      "epoch": 1.8098510882016037,
      "grad_norm": 0.534919240445055,
      "learning_rate": 5e-06,
      "loss": 0.3763,
      "step": 3160
    },
    {
      "epoch": 1.8155784650630011,
      "grad_norm": 0.5146135891570391,
      "learning_rate": 5e-06,
      "loss": 0.3806,
      "step": 3170
    },
    {
      "epoch": 1.8213058419243986,
      "grad_norm": 0.505605657209453,
      "learning_rate": 5e-06,
      "loss": 0.3766,
      "step": 3180
    },
    {
      "epoch": 1.827033218785796,
      "grad_norm": 0.5239153217495089,
      "learning_rate": 5e-06,
      "loss": 0.3659,
      "step": 3190
    },
    {
      "epoch": 1.8327605956471937,
      "grad_norm": 0.9509826091058188,
      "learning_rate": 5e-06,
      "loss": 0.3792,
      "step": 3200
    },
    {
      "epoch": 1.8384879725085912,
      "grad_norm": 0.5059729711091226,
      "learning_rate": 5e-06,
      "loss": 0.3721,
      "step": 3210
    },
    {
      "epoch": 1.8442153493699887,
      "grad_norm": 0.5245359643059282,
      "learning_rate": 5e-06,
      "loss": 0.379,
      "step": 3220
    },
    {
      "epoch": 1.8499427262313861,
      "grad_norm": 0.5050698399570109,
      "learning_rate": 5e-06,
      "loss": 0.3739,
      "step": 3230
    },
    {
      "epoch": 1.8556701030927836,
      "grad_norm": 0.5065697206422057,
      "learning_rate": 5e-06,
      "loss": 0.3849,
      "step": 3240
    },
    {
      "epoch": 1.861397479954181,
      "grad_norm": 0.5209454651612013,
      "learning_rate": 5e-06,
      "loss": 0.3707,
      "step": 3250
    },
    {
      "epoch": 1.8671248568155785,
      "grad_norm": 0.5430271570033155,
      "learning_rate": 5e-06,
      "loss": 0.3696,
      "step": 3260
    },
    {
      "epoch": 1.872852233676976,
      "grad_norm": 0.5281297299023378,
      "learning_rate": 5e-06,
      "loss": 0.3712,
      "step": 3270
    },
    {
      "epoch": 1.8785796105383734,
      "grad_norm": 0.5292724874313788,
      "learning_rate": 5e-06,
      "loss": 0.3726,
      "step": 3280
    },
    {
      "epoch": 1.8843069873997709,
      "grad_norm": 0.5350000049801282,
      "learning_rate": 5e-06,
      "loss": 0.3723,
      "step": 3290
    },
    {
      "epoch": 1.8900343642611683,
      "grad_norm": 0.5208368353063128,
      "learning_rate": 5e-06,
      "loss": 0.3677,
      "step": 3300
    },
    {
      "epoch": 1.8957617411225658,
      "grad_norm": 0.5279268625494672,
      "learning_rate": 5e-06,
      "loss": 0.3753,
      "step": 3310
    },
    {
      "epoch": 1.9014891179839633,
      "grad_norm": 0.5144354886651064,
      "learning_rate": 5e-06,
      "loss": 0.3741,
      "step": 3320
    },
    {
      "epoch": 1.9072164948453607,
      "grad_norm": 0.5082140336402234,
      "learning_rate": 5e-06,
      "loss": 0.3635,
      "step": 3330
    },
    {
      "epoch": 1.9129438717067582,
      "grad_norm": 0.5252002760881638,
      "learning_rate": 5e-06,
      "loss": 0.3787,
      "step": 3340
    },
    {
      "epoch": 1.9186712485681556,
      "grad_norm": 0.5070002443248173,
      "learning_rate": 5e-06,
      "loss": 0.3718,
      "step": 3350
    },
    {
      "epoch": 1.9243986254295533,
      "grad_norm": 0.5316148384707632,
      "learning_rate": 5e-06,
      "loss": 0.3664,
      "step": 3360
    },
    {
      "epoch": 1.9301260022909508,
      "grad_norm": 0.5193287345279981,
      "learning_rate": 5e-06,
      "loss": 0.3684,
      "step": 3370
    },
    {
      "epoch": 1.9358533791523482,
      "grad_norm": 0.506552724781655,
      "learning_rate": 5e-06,
      "loss": 0.372,
      "step": 3380
    },
    {
      "epoch": 1.9415807560137457,
      "grad_norm": 0.571442610460597,
      "learning_rate": 5e-06,
      "loss": 0.379,
      "step": 3390
    },
    {
      "epoch": 1.9473081328751431,
      "grad_norm": 0.5688327247671221,
      "learning_rate": 5e-06,
      "loss": 0.3687,
      "step": 3400
    },
    {
      "epoch": 1.9530355097365406,
      "grad_norm": 0.5042217231517675,
      "learning_rate": 5e-06,
      "loss": 0.3738,
      "step": 3410
    },
    {
      "epoch": 1.9587628865979383,
      "grad_norm": 0.5321243972550566,
      "learning_rate": 5e-06,
      "loss": 0.37,
      "step": 3420
    },
    {
      "epoch": 1.9644902634593358,
      "grad_norm": 0.5188108633782941,
      "learning_rate": 5e-06,
      "loss": 0.3766,
      "step": 3430
    },
    {
      "epoch": 1.9702176403207332,
      "grad_norm": 0.49989426974744805,
      "learning_rate": 5e-06,
      "loss": 0.3703,
      "step": 3440
    },
    {
      "epoch": 1.9759450171821307,
      "grad_norm": 0.5428089945263836,
      "learning_rate": 5e-06,
      "loss": 0.3689,
      "step": 3450
    },
    {
      "epoch": 1.9816723940435281,
      "grad_norm": 0.5404439065398382,
      "learning_rate": 5e-06,
      "loss": 0.3729,
      "step": 3460
    },
    {
      "epoch": 1.9873997709049256,
      "grad_norm": 0.5410379700662915,
      "learning_rate": 5e-06,
      "loss": 0.3726,
      "step": 3470
    },
    {
      "epoch": 1.993127147766323,
      "grad_norm": 0.5225961306378605,
      "learning_rate": 5e-06,
      "loss": 0.3782,
      "step": 3480
    },
    {
      "epoch": 1.9988545246277205,
      "grad_norm": 0.5535365053498117,
      "learning_rate": 5e-06,
      "loss": 0.3757,
      "step": 3490
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.42091062664985657,
      "eval_runtime": 444.2503,
      "eval_samples_per_second": 26.476,
      "eval_steps_per_second": 0.414,
      "step": 3492
    },
    {
      "epoch": 2.004581901489118,
      "grad_norm": 0.5715227391793876,
      "learning_rate": 5e-06,
      "loss": 0.3266,
      "step": 3500
    },
    {
      "epoch": 2.0103092783505154,
      "grad_norm": 0.5115288441622716,
      "learning_rate": 5e-06,
      "loss": 0.3123,
      "step": 3510
    },
    {
      "epoch": 2.016036655211913,
      "grad_norm": 0.5114186476378918,
      "learning_rate": 5e-06,
      "loss": 0.3064,
      "step": 3520
    },
    {
      "epoch": 2.0217640320733103,
      "grad_norm": 0.5174383277753903,
      "learning_rate": 5e-06,
      "loss": 0.3115,
      "step": 3530
    },
    {
      "epoch": 2.027491408934708,
      "grad_norm": 0.5180676439991899,
      "learning_rate": 5e-06,
      "loss": 0.3092,
      "step": 3540
    },
    {
      "epoch": 2.0332187857961053,
      "grad_norm": 0.5521743990569499,
      "learning_rate": 5e-06,
      "loss": 0.3154,
      "step": 3550
    },
    {
      "epoch": 2.0389461626575027,
      "grad_norm": 0.5325564269068459,
      "learning_rate": 5e-06,
      "loss": 0.3139,
      "step": 3560
    },
    {
      "epoch": 2.0446735395189,
      "grad_norm": 0.5307121051642008,
      "learning_rate": 5e-06,
      "loss": 0.3143,
      "step": 3570
    },
    {
      "epoch": 2.0504009163802976,
      "grad_norm": 0.5039022580188123,
      "learning_rate": 5e-06,
      "loss": 0.3139,
      "step": 3580
    },
    {
      "epoch": 2.056128293241695,
      "grad_norm": 0.5386599812235912,
      "learning_rate": 5e-06,
      "loss": 0.3274,
      "step": 3590
    },
    {
      "epoch": 2.0618556701030926,
      "grad_norm": 0.5510122632798343,
      "learning_rate": 5e-06,
      "loss": 0.3218,
      "step": 3600
    },
    {
      "epoch": 2.0675830469644905,
      "grad_norm": 0.5314262243630897,
      "learning_rate": 5e-06,
      "loss": 0.3126,
      "step": 3610
    },
    {
      "epoch": 2.073310423825888,
      "grad_norm": 0.4939184680389998,
      "learning_rate": 5e-06,
      "loss": 0.313,
      "step": 3620
    },
    {
      "epoch": 2.0790378006872854,
      "grad_norm": 0.5312396764234006,
      "learning_rate": 5e-06,
      "loss": 0.3105,
      "step": 3630
    },
    {
      "epoch": 2.084765177548683,
      "grad_norm": 0.4876119734890598,
      "learning_rate": 5e-06,
      "loss": 0.3124,
      "step": 3640
    },
    {
      "epoch": 2.0904925544100803,
      "grad_norm": 0.5103227941442228,
      "learning_rate": 5e-06,
      "loss": 0.3242,
      "step": 3650
    },
    {
      "epoch": 2.0962199312714778,
      "grad_norm": 0.5042566969256604,
      "learning_rate": 5e-06,
      "loss": 0.3184,
      "step": 3660
    },
    {
      "epoch": 2.1019473081328752,
      "grad_norm": 0.5141761625174609,
      "learning_rate": 5e-06,
      "loss": 0.3143,
      "step": 3670
    },
    {
      "epoch": 2.1076746849942727,
      "grad_norm": 0.5382583222094659,
      "learning_rate": 5e-06,
      "loss": 0.3166,
      "step": 3680
    },
    {
      "epoch": 2.11340206185567,
      "grad_norm": 0.4804864664629375,
      "learning_rate": 5e-06,
      "loss": 0.3122,
      "step": 3690
    },
    {
      "epoch": 2.1191294387170676,
      "grad_norm": 0.5165262098916743,
      "learning_rate": 5e-06,
      "loss": 0.3135,
      "step": 3700
    },
    {
      "epoch": 2.124856815578465,
      "grad_norm": 0.5253973489118585,
      "learning_rate": 5e-06,
      "loss": 0.3103,
      "step": 3710
    },
    {
      "epoch": 2.1305841924398625,
      "grad_norm": 0.5153233857956726,
      "learning_rate": 5e-06,
      "loss": 0.319,
      "step": 3720
    },
    {
      "epoch": 2.13631156930126,
      "grad_norm": 0.5024911676484453,
      "learning_rate": 5e-06,
      "loss": 0.3153,
      "step": 3730
    },
    {
      "epoch": 2.1420389461626574,
      "grad_norm": 0.48649497978161954,
      "learning_rate": 5e-06,
      "loss": 0.3125,
      "step": 3740
    },
    {
      "epoch": 2.147766323024055,
      "grad_norm": 0.5398906998916857,
      "learning_rate": 5e-06,
      "loss": 0.3207,
      "step": 3750
    },
    {
      "epoch": 2.1534936998854524,
      "grad_norm": 0.5130668071668567,
      "learning_rate": 5e-06,
      "loss": 0.3141,
      "step": 3760
    },
    {
      "epoch": 2.15922107674685,
      "grad_norm": 0.5518874676047263,
      "learning_rate": 5e-06,
      "loss": 0.3184,
      "step": 3770
    },
    {
      "epoch": 2.1649484536082473,
      "grad_norm": 0.5160046944915151,
      "learning_rate": 5e-06,
      "loss": 0.3172,
      "step": 3780
    },
    {
      "epoch": 2.1706758304696447,
      "grad_norm": 0.48834912906902556,
      "learning_rate": 5e-06,
      "loss": 0.3234,
      "step": 3790
    },
    {
      "epoch": 2.176403207331042,
      "grad_norm": 0.5171110867595475,
      "learning_rate": 5e-06,
      "loss": 0.3151,
      "step": 3800
    },
    {
      "epoch": 2.1821305841924397,
      "grad_norm": 0.5180778501865309,
      "learning_rate": 5e-06,
      "loss": 0.3157,
      "step": 3810
    },
    {
      "epoch": 2.1878579610538376,
      "grad_norm": 0.5068334930400914,
      "learning_rate": 5e-06,
      "loss": 0.3126,
      "step": 3820
    },
    {
      "epoch": 2.193585337915235,
      "grad_norm": 0.5383783523767623,
      "learning_rate": 5e-06,
      "loss": 0.313,
      "step": 3830
    },
    {
      "epoch": 2.1993127147766325,
      "grad_norm": 0.5231488887255323,
      "learning_rate": 5e-06,
      "loss": 0.3199,
      "step": 3840
    },
    {
      "epoch": 2.20504009163803,
      "grad_norm": 0.5130199004591844,
      "learning_rate": 5e-06,
      "loss": 0.3168,
      "step": 3850
    },
    {
      "epoch": 2.2107674684994274,
      "grad_norm": 0.5482589908800253,
      "learning_rate": 5e-06,
      "loss": 0.3223,
      "step": 3860
    },
    {
      "epoch": 2.216494845360825,
      "grad_norm": 0.53120848952022,
      "learning_rate": 5e-06,
      "loss": 0.321,
      "step": 3870
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.5071013057492006,
      "learning_rate": 5e-06,
      "loss": 0.3203,
      "step": 3880
    },
    {
      "epoch": 2.22794959908362,
      "grad_norm": 0.531266327465737,
      "learning_rate": 5e-06,
      "loss": 0.324,
      "step": 3890
    },
    {
      "epoch": 2.2336769759450172,
      "grad_norm": 0.507052566002429,
      "learning_rate": 5e-06,
      "loss": 0.3137,
      "step": 3900
    },
    {
      "epoch": 2.2394043528064147,
      "grad_norm": 0.5271639744898703,
      "learning_rate": 5e-06,
      "loss": 0.3159,
      "step": 3910
    },
    {
      "epoch": 2.245131729667812,
      "grad_norm": 0.5388005656911042,
      "learning_rate": 5e-06,
      "loss": 0.3228,
      "step": 3920
    },
    {
      "epoch": 2.2508591065292096,
      "grad_norm": 0.5060194168439476,
      "learning_rate": 5e-06,
      "loss": 0.3124,
      "step": 3930
    },
    {
      "epoch": 2.256586483390607,
      "grad_norm": 0.48546648398689385,
      "learning_rate": 5e-06,
      "loss": 0.3136,
      "step": 3940
    },
    {
      "epoch": 2.2623138602520045,
      "grad_norm": 0.5076224304840579,
      "learning_rate": 5e-06,
      "loss": 0.3167,
      "step": 3950
    },
    {
      "epoch": 2.268041237113402,
      "grad_norm": 0.5296050756375043,
      "learning_rate": 5e-06,
      "loss": 0.3204,
      "step": 3960
    },
    {
      "epoch": 2.2737686139747995,
      "grad_norm": 0.5177826901430715,
      "learning_rate": 5e-06,
      "loss": 0.3232,
      "step": 3970
    },
    {
      "epoch": 2.279495990836197,
      "grad_norm": 0.5132840661554632,
      "learning_rate": 5e-06,
      "loss": 0.3158,
      "step": 3980
    },
    {
      "epoch": 2.2852233676975944,
      "grad_norm": 0.543452264339846,
      "learning_rate": 5e-06,
      "loss": 0.3192,
      "step": 3990
    },
    {
      "epoch": 2.290950744558992,
      "grad_norm": 0.5139552493291161,
      "learning_rate": 5e-06,
      "loss": 0.3209,
      "step": 4000
    },
    {
      "epoch": 2.2966781214203893,
      "grad_norm": 0.4969492673393391,
      "learning_rate": 5e-06,
      "loss": 0.3153,
      "step": 4010
    },
    {
      "epoch": 2.3024054982817868,
      "grad_norm": 0.5143134020150514,
      "learning_rate": 5e-06,
      "loss": 0.3201,
      "step": 4020
    },
    {
      "epoch": 2.308132875143184,
      "grad_norm": 0.4939050512552889,
      "learning_rate": 5e-06,
      "loss": 0.3106,
      "step": 4030
    },
    {
      "epoch": 2.3138602520045817,
      "grad_norm": 0.4886273394938644,
      "learning_rate": 5e-06,
      "loss": 0.3225,
      "step": 4040
    },
    {
      "epoch": 2.319587628865979,
      "grad_norm": 0.5380016012501289,
      "learning_rate": 5e-06,
      "loss": 0.3183,
      "step": 4050
    },
    {
      "epoch": 2.325315005727377,
      "grad_norm": 0.5160375481982886,
      "learning_rate": 5e-06,
      "loss": 0.3127,
      "step": 4060
    },
    {
      "epoch": 2.3310423825887745,
      "grad_norm": 0.5217824879654166,
      "learning_rate": 5e-06,
      "loss": 0.3168,
      "step": 4070
    },
    {
      "epoch": 2.336769759450172,
      "grad_norm": 0.4987920921308887,
      "learning_rate": 5e-06,
      "loss": 0.3243,
      "step": 4080
    },
    {
      "epoch": 2.3424971363115694,
      "grad_norm": 0.540718540329614,
      "learning_rate": 5e-06,
      "loss": 0.3267,
      "step": 4090
    },
    {
      "epoch": 2.348224513172967,
      "grad_norm": 0.4928104897116783,
      "learning_rate": 5e-06,
      "loss": 0.3154,
      "step": 4100
    },
    {
      "epoch": 2.3539518900343643,
      "grad_norm": 0.5332905023045372,
      "learning_rate": 5e-06,
      "loss": 0.3294,
      "step": 4110
    },
    {
      "epoch": 2.359679266895762,
      "grad_norm": 0.5190878954413758,
      "learning_rate": 5e-06,
      "loss": 0.3135,
      "step": 4120
    },
    {
      "epoch": 2.3654066437571593,
      "grad_norm": 0.49770844545756987,
      "learning_rate": 5e-06,
      "loss": 0.3217,
      "step": 4130
    },
    {
      "epoch": 2.3711340206185567,
      "grad_norm": 0.5158902735624868,
      "learning_rate": 5e-06,
      "loss": 0.3182,
      "step": 4140
    },
    {
      "epoch": 2.376861397479954,
      "grad_norm": 0.5258004488461884,
      "learning_rate": 5e-06,
      "loss": 0.3211,
      "step": 4150
    },
    {
      "epoch": 2.3825887743413516,
      "grad_norm": 0.5006246348277473,
      "learning_rate": 5e-06,
      "loss": 0.311,
      "step": 4160
    },
    {
      "epoch": 2.388316151202749,
      "grad_norm": 0.5152750751947476,
      "learning_rate": 5e-06,
      "loss": 0.3219,
      "step": 4170
    },
    {
      "epoch": 2.3940435280641466,
      "grad_norm": 0.52068469477404,
      "learning_rate": 5e-06,
      "loss": 0.3164,
      "step": 4180
    },
    {
      "epoch": 2.399770904925544,
      "grad_norm": 0.5323767239444738,
      "learning_rate": 5e-06,
      "loss": 0.3208,
      "step": 4190
    },
    {
      "epoch": 2.4054982817869415,
      "grad_norm": 0.5003101909687979,
      "learning_rate": 5e-06,
      "loss": 0.3228,
      "step": 4200
    },
    {
      "epoch": 2.411225658648339,
      "grad_norm": 0.5111739731033206,
      "learning_rate": 5e-06,
      "loss": 0.3271,
      "step": 4210
    },
    {
      "epoch": 2.4169530355097364,
      "grad_norm": 0.5131151305511839,
      "learning_rate": 5e-06,
      "loss": 0.3147,
      "step": 4220
    },
    {
      "epoch": 2.422680412371134,
      "grad_norm": 0.5163340601407542,
      "learning_rate": 5e-06,
      "loss": 0.3194,
      "step": 4230
    },
    {
      "epoch": 2.4284077892325313,
      "grad_norm": 0.4958432705018228,
      "learning_rate": 5e-06,
      "loss": 0.3188,
      "step": 4240
    },
    {
      "epoch": 2.434135166093929,
      "grad_norm": 0.5468740104263736,
      "learning_rate": 5e-06,
      "loss": 0.3249,
      "step": 4250
    },
    {
      "epoch": 2.4398625429553267,
      "grad_norm": 0.5410705485869296,
      "learning_rate": 5e-06,
      "loss": 0.3278,
      "step": 4260
    },
    {
      "epoch": 2.445589919816724,
      "grad_norm": 0.5368673433466151,
      "learning_rate": 5e-06,
      "loss": 0.3076,
      "step": 4270
    },
    {
      "epoch": 2.4513172966781216,
      "grad_norm": 0.5128839768898594,
      "learning_rate": 5e-06,
      "loss": 0.3184,
      "step": 4280
    },
    {
      "epoch": 2.457044673539519,
      "grad_norm": 0.5068335080164907,
      "learning_rate": 5e-06,
      "loss": 0.3169,
      "step": 4290
    },
    {
      "epoch": 2.4627720504009165,
      "grad_norm": 0.4997641306590389,
      "learning_rate": 5e-06,
      "loss": 0.3182,
      "step": 4300
    },
    {
      "epoch": 2.468499427262314,
      "grad_norm": 0.5191935789555608,
      "learning_rate": 5e-06,
      "loss": 0.3216,
      "step": 4310
    },
    {
      "epoch": 2.4742268041237114,
      "grad_norm": 0.5256779474029062,
      "learning_rate": 5e-06,
      "loss": 0.3233,
      "step": 4320
    },
    {
      "epoch": 2.479954180985109,
      "grad_norm": 0.4865715340962184,
      "learning_rate": 5e-06,
      "loss": 0.3204,
      "step": 4330
    },
    {
      "epoch": 2.4856815578465064,
      "grad_norm": 0.526248237049682,
      "learning_rate": 5e-06,
      "loss": 0.3205,
      "step": 4340
    },
    {
      "epoch": 2.491408934707904,
      "grad_norm": 0.5130011897003942,
      "learning_rate": 5e-06,
      "loss": 0.3259,
      "step": 4350
    },
    {
      "epoch": 2.4971363115693013,
      "grad_norm": 0.5162374829270555,
      "learning_rate": 5e-06,
      "loss": 0.3195,
      "step": 4360
    },
    {
      "epoch": 2.5028636884306987,
      "grad_norm": 0.5014077765674124,
      "learning_rate": 5e-06,
      "loss": 0.3228,
      "step": 4370
    },
    {
      "epoch": 2.508591065292096,
      "grad_norm": 0.5026646925273252,
      "learning_rate": 5e-06,
      "loss": 0.3207,
      "step": 4380
    },
    {
      "epoch": 2.5143184421534936,
      "grad_norm": 0.489097601061641,
      "learning_rate": 5e-06,
      "loss": 0.3176,
      "step": 4390
    },
    {
      "epoch": 2.520045819014891,
      "grad_norm": 0.5078735499696359,
      "learning_rate": 5e-06,
      "loss": 0.3224,
      "step": 4400
    },
    {
      "epoch": 2.5257731958762886,
      "grad_norm": 0.5271652940063883,
      "learning_rate": 5e-06,
      "loss": 0.323,
      "step": 4410
    },
    {
      "epoch": 2.531500572737686,
      "grad_norm": 0.507078592732912,
      "learning_rate": 5e-06,
      "loss": 0.3203,
      "step": 4420
    },
    {
      "epoch": 2.5372279495990835,
      "grad_norm": 0.5337080175384938,
      "learning_rate": 5e-06,
      "loss": 0.3228,
      "step": 4430
    },
    {
      "epoch": 2.542955326460481,
      "grad_norm": 0.50176908036031,
      "learning_rate": 5e-06,
      "loss": 0.3202,
      "step": 4440
    },
    {
      "epoch": 2.5486827033218784,
      "grad_norm": 0.5120277803126527,
      "learning_rate": 5e-06,
      "loss": 0.3197,
      "step": 4450
    },
    {
      "epoch": 2.554410080183276,
      "grad_norm": 0.5366578152814068,
      "learning_rate": 5e-06,
      "loss": 0.3244,
      "step": 4460
    },
    {
      "epoch": 2.5601374570446733,
      "grad_norm": 0.522159840101665,
      "learning_rate": 5e-06,
      "loss": 0.3185,
      "step": 4470
    },
    {
      "epoch": 2.565864833906071,
      "grad_norm": 0.48196371586373893,
      "learning_rate": 5e-06,
      "loss": 0.3173,
      "step": 4480
    },
    {
      "epoch": 2.5715922107674682,
      "grad_norm": 0.5255302877210084,
      "learning_rate": 5e-06,
      "loss": 0.3278,
      "step": 4490
    },
    {
      "epoch": 2.5773195876288657,
      "grad_norm": 0.5089869190508646,
      "learning_rate": 5e-06,
      "loss": 0.3198,
      "step": 4500
    },
    {
      "epoch": 2.5830469644902636,
      "grad_norm": 0.530959409102577,
      "learning_rate": 5e-06,
      "loss": 0.3304,
      "step": 4510
    },
    {
      "epoch": 2.588774341351661,
      "grad_norm": 0.5531110752439229,
      "learning_rate": 5e-06,
      "loss": 0.3215,
      "step": 4520
    },
    {
      "epoch": 2.5945017182130585,
      "grad_norm": 0.5260126942389958,
      "learning_rate": 5e-06,
      "loss": 0.316,
      "step": 4530
    },
    {
      "epoch": 2.600229095074456,
      "grad_norm": 0.5160707506662007,
      "learning_rate": 5e-06,
      "loss": 0.32,
      "step": 4540
    },
    {
      "epoch": 2.6059564719358534,
      "grad_norm": 0.5410876892071711,
      "learning_rate": 5e-06,
      "loss": 0.324,
      "step": 4550
    },
    {
      "epoch": 2.611683848797251,
      "grad_norm": 0.5149983976318435,
      "learning_rate": 5e-06,
      "loss": 0.3168,
      "step": 4560
    },
    {
      "epoch": 2.6174112256586484,
      "grad_norm": 0.5066882681673974,
      "learning_rate": 5e-06,
      "loss": 0.3164,
      "step": 4570
    },
    {
      "epoch": 2.623138602520046,
      "grad_norm": 0.5033788230089447,
      "learning_rate": 5e-06,
      "loss": 0.3163,
      "step": 4580
    },
    {
      "epoch": 2.6288659793814433,
      "grad_norm": 0.5247195850361597,
      "learning_rate": 5e-06,
      "loss": 0.3175,
      "step": 4590
    },
    {
      "epoch": 2.6345933562428407,
      "grad_norm": 0.5282100918029693,
      "learning_rate": 5e-06,
      "loss": 0.3272,
      "step": 4600
    },
    {
      "epoch": 2.640320733104238,
      "grad_norm": 0.52717520655256,
      "learning_rate": 5e-06,
      "loss": 0.3203,
      "step": 4610
    },
    {
      "epoch": 2.6460481099656357,
      "grad_norm": 0.5022071775606154,
      "learning_rate": 5e-06,
      "loss": 0.3244,
      "step": 4620
    },
    {
      "epoch": 2.651775486827033,
      "grad_norm": 0.5211994501987708,
      "learning_rate": 5e-06,
      "loss": 0.3131,
      "step": 4630
    },
    {
      "epoch": 2.6575028636884306,
      "grad_norm": 0.5124524380744641,
      "learning_rate": 5e-06,
      "loss": 0.3253,
      "step": 4640
    },
    {
      "epoch": 2.663230240549828,
      "grad_norm": 0.5060206463796821,
      "learning_rate": 5e-06,
      "loss": 0.3173,
      "step": 4650
    },
    {
      "epoch": 2.6689576174112255,
      "grad_norm": 0.5162650933726732,
      "learning_rate": 5e-06,
      "loss": 0.3198,
      "step": 4660
    },
    {
      "epoch": 2.6746849942726234,
      "grad_norm": 0.524408740926564,
      "learning_rate": 5e-06,
      "loss": 0.3302,
      "step": 4670
    },
    {
      "epoch": 2.680412371134021,
      "grad_norm": 0.5146510600522526,
      "learning_rate": 5e-06,
      "loss": 0.3234,
      "step": 4680
    },
    {
      "epoch": 2.6861397479954183,
      "grad_norm": 0.5092928068009863,
      "learning_rate": 5e-06,
      "loss": 0.322,
      "step": 4690
    },
    {
      "epoch": 2.691867124856816,
      "grad_norm": 0.5757942628503555,
      "learning_rate": 5e-06,
      "loss": 0.3117,
      "step": 4700
    },
    {
      "epoch": 2.6975945017182132,
      "grad_norm": 0.4889008213010037,
      "learning_rate": 5e-06,
      "loss": 0.3231,
      "step": 4710
    },
    {
      "epoch": 2.7033218785796107,
      "grad_norm": 0.5302974244163768,
      "learning_rate": 5e-06,
      "loss": 0.3178,
      "step": 4720
    },
    {
      "epoch": 2.709049255441008,
      "grad_norm": 0.52479839402123,
      "learning_rate": 5e-06,
      "loss": 0.3217,
      "step": 4730
    },
    {
      "epoch": 2.7147766323024056,
      "grad_norm": 0.5039321357679906,
      "learning_rate": 5e-06,
      "loss": 0.3257,
      "step": 4740
    },
    {
      "epoch": 2.720504009163803,
      "grad_norm": 0.5101102471452367,
      "learning_rate": 5e-06,
      "loss": 0.3222,
      "step": 4750
    },
    {
      "epoch": 2.7262313860252005,
      "grad_norm": 0.5430506058097,
      "learning_rate": 5e-06,
      "loss": 0.319,
      "step": 4760
    },
    {
      "epoch": 2.731958762886598,
      "grad_norm": 0.5355617323855306,
      "learning_rate": 5e-06,
      "loss": 0.3269,
      "step": 4770
    },
    {
      "epoch": 2.7376861397479955,
      "grad_norm": 0.4986561926249133,
      "learning_rate": 5e-06,
      "loss": 0.3149,
      "step": 4780
    },
    {
      "epoch": 2.743413516609393,
      "grad_norm": 0.5053547200537913,
      "learning_rate": 5e-06,
      "loss": 0.3308,
      "step": 4790
    },
    {
      "epoch": 2.7491408934707904,
      "grad_norm": 0.48632254517347523,
      "learning_rate": 5e-06,
      "loss": 0.3185,
      "step": 4800
    },
    {
      "epoch": 2.754868270332188,
      "grad_norm": 0.486358386605019,
      "learning_rate": 5e-06,
      "loss": 0.3103,
      "step": 4810
    },
    {
      "epoch": 2.7605956471935853,
      "grad_norm": 0.5374470855558577,
      "learning_rate": 5e-06,
      "loss": 0.3298,
      "step": 4820
    },
    {
      "epoch": 2.7663230240549828,
      "grad_norm": 0.5153628917876052,
      "learning_rate": 5e-06,
      "loss": 0.3233,
      "step": 4830
    },
    {
      "epoch": 2.77205040091638,
      "grad_norm": 0.5054154722646462,
      "learning_rate": 5e-06,
      "loss": 0.3275,
      "step": 4840
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.5149761064711591,
      "learning_rate": 5e-06,
      "loss": 0.317,
      "step": 4850
    },
    {
      "epoch": 2.783505154639175,
      "grad_norm": 0.5148931201825926,
      "learning_rate": 5e-06,
      "loss": 0.3233,
      "step": 4860
    },
    {
      "epoch": 2.7892325315005726,
      "grad_norm": 0.5191662793295072,
      "learning_rate": 5e-06,
      "loss": 0.3184,
      "step": 4870
    },
    {
      "epoch": 2.79495990836197,
      "grad_norm": 0.5675073234527983,
      "learning_rate": 5e-06,
      "loss": 0.3251,
      "step": 4880
    },
    {
      "epoch": 2.8006872852233675,
      "grad_norm": 0.5038525244128136,
      "learning_rate": 5e-06,
      "loss": 0.3232,
      "step": 4890
    },
    {
      "epoch": 2.806414662084765,
      "grad_norm": 0.4918634907852925,
      "learning_rate": 5e-06,
      "loss": 0.3175,
      "step": 4900
    },
    {
      "epoch": 2.8121420389461624,
      "grad_norm": 0.5284779685936459,
      "learning_rate": 5e-06,
      "loss": 0.3208,
      "step": 4910
    },
    {
      "epoch": 2.81786941580756,
      "grad_norm": 0.5108078870269828,
      "learning_rate": 5e-06,
      "loss": 0.3272,
      "step": 4920
    },
    {
      "epoch": 2.8235967926689574,
      "grad_norm": 0.4902747564813571,
      "learning_rate": 5e-06,
      "loss": 0.3257,
      "step": 4930
    },
    {
      "epoch": 2.8293241695303553,
      "grad_norm": 0.5030822928949704,
      "learning_rate": 5e-06,
      "loss": 0.32,
      "step": 4940
    },
    {
      "epoch": 2.8350515463917527,
      "grad_norm": 0.5272339203419648,
      "learning_rate": 5e-06,
      "loss": 0.3288,
      "step": 4950
    },
    {
      "epoch": 2.84077892325315,
      "grad_norm": 0.5340619443879843,
      "learning_rate": 5e-06,
      "loss": 0.3277,
      "step": 4960
    },
    {
      "epoch": 2.8465063001145476,
      "grad_norm": 0.5139875993534403,
      "learning_rate": 5e-06,
      "loss": 0.3167,
      "step": 4970
    },
    {
      "epoch": 2.852233676975945,
      "grad_norm": 0.5197113622112696,
      "learning_rate": 5e-06,
      "loss": 0.3221,
      "step": 4980
    },
    {
      "epoch": 2.8579610538373426,
      "grad_norm": 0.4932910708419312,
      "learning_rate": 5e-06,
      "loss": 0.319,
      "step": 4990
    },
    {
      "epoch": 2.86368843069874,
      "grad_norm": 0.48961874681440903,
      "learning_rate": 5e-06,
      "loss": 0.3216,
      "step": 5000
    },
    {
      "epoch": 2.8694158075601375,
      "grad_norm": 0.5306801859906654,
      "learning_rate": 5e-06,
      "loss": 0.3294,
      "step": 5010
    },
    {
      "epoch": 2.875143184421535,
      "grad_norm": 0.5245022306778138,
      "learning_rate": 5e-06,
      "loss": 0.3238,
      "step": 5020
    },
    {
      "epoch": 2.8808705612829324,
      "grad_norm": 0.5294236720116192,
      "learning_rate": 5e-06,
      "loss": 0.3172,
      "step": 5030
    },
    {
      "epoch": 2.88659793814433,
      "grad_norm": 0.49805002487820915,
      "learning_rate": 5e-06,
      "loss": 0.3154,
      "step": 5040
    },
    {
      "epoch": 2.8923253150057273,
      "grad_norm": 0.5118585995041246,
      "learning_rate": 5e-06,
      "loss": 0.3264,
      "step": 5050
    },
    {
      "epoch": 2.8980526918671248,
      "grad_norm": 0.4778628715181792,
      "learning_rate": 5e-06,
      "loss": 0.3215,
      "step": 5060
    },
    {
      "epoch": 2.9037800687285222,
      "grad_norm": 0.502141421148047,
      "learning_rate": 5e-06,
      "loss": 0.3195,
      "step": 5070
    },
    {
      "epoch": 2.9095074455899197,
      "grad_norm": 0.5216144225169003,
      "learning_rate": 5e-06,
      "loss": 0.322,
      "step": 5080
    },
    {
      "epoch": 2.915234822451317,
      "grad_norm": 0.5061317551186302,
      "learning_rate": 5e-06,
      "loss": 0.3181,
      "step": 5090
    },
    {
      "epoch": 2.9209621993127146,
      "grad_norm": 0.5258296449599809,
      "learning_rate": 5e-06,
      "loss": 0.3196,
      "step": 5100
    },
    {
      "epoch": 2.9266895761741125,
      "grad_norm": 0.5205129759914494,
      "learning_rate": 5e-06,
      "loss": 0.3298,
      "step": 5110
    },
    {
      "epoch": 2.93241695303551,
      "grad_norm": 0.5441889203090563,
      "learning_rate": 5e-06,
      "loss": 0.3247,
      "step": 5120
    },
    {
      "epoch": 2.9381443298969074,
      "grad_norm": 0.5039522474134192,
      "learning_rate": 5e-06,
      "loss": 0.3181,
      "step": 5130
    },
    {
      "epoch": 2.943871706758305,
      "grad_norm": 0.5232711939948509,
      "learning_rate": 5e-06,
      "loss": 0.3196,
      "step": 5140
    },
    {
      "epoch": 2.9495990836197024,
      "grad_norm": 0.5119250727874893,
      "learning_rate": 5e-06,
      "loss": 0.3131,
      "step": 5150
    },
    {
      "epoch": 2.9553264604811,
      "grad_norm": 0.5050585849003654,
      "learning_rate": 5e-06,
      "loss": 0.3248,
      "step": 5160
    },
    {
      "epoch": 2.9610538373424973,
      "grad_norm": 0.5164002352592711,
      "learning_rate": 5e-06,
      "loss": 0.32,
      "step": 5170
    },
    {
      "epoch": 2.9667812142038947,
      "grad_norm": 0.5133798037929324,
      "learning_rate": 5e-06,
      "loss": 0.3185,
      "step": 5180
    },
    {
      "epoch": 2.972508591065292,
      "grad_norm": 0.5131629955273521,
      "learning_rate": 5e-06,
      "loss": 0.3169,
      "step": 5190
    },
    {
      "epoch": 2.9782359679266897,
      "grad_norm": 0.5354127604897547,
      "learning_rate": 5e-06,
      "loss": 0.3265,
      "step": 5200
    },
    {
      "epoch": 2.983963344788087,
      "grad_norm": 0.486875785468725,
      "learning_rate": 5e-06,
      "loss": 0.3185,
      "step": 5210
    },
    {
      "epoch": 2.9896907216494846,
      "grad_norm": 0.524969500670149,
      "learning_rate": 5e-06,
      "loss": 0.3161,
      "step": 5220
    },
    {
      "epoch": 2.995418098510882,
      "grad_norm": 0.5109832301608017,
      "learning_rate": 5e-06,
      "loss": 0.3268,
      "step": 5230
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.42792943120002747,
      "eval_runtime": 443.2838,
      "eval_samples_per_second": 26.534,
      "eval_steps_per_second": 0.415,
      "step": 5238
    },
    {
      "epoch": 3.0,
      "step": 5238,
      "total_flos": 2745782658662400.0,
      "train_loss": 0.3825414228548572,
      "train_runtime": 71090.721,
      "train_samples_per_second": 9.431,
      "train_steps_per_second": 0.074
    }
  ],
  "logging_steps": 10,
  "max_steps": 5238,
  "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": 2745782658662400.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}