{
  "best_metric": 1.184820294380188,
  "best_model_checkpoint": "./mamba-790m-phase2/checkpoint-2850",
  "epoch": 1.2324324324324325,
  "eval_steps": 50,
  "global_step": 2850,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008648648648648649,
      "grad_norm": 0.13316184282302856,
      "learning_rate": 2.158273381294964e-06,
      "loss": 1.3731,
      "step": 20
    },
    {
      "epoch": 0.017297297297297298,
      "grad_norm": 0.12288903445005417,
      "learning_rate": 4.316546762589928e-06,
      "loss": 1.5117,
      "step": 40
    },
    {
      "epoch": 0.021621621621621623,
      "eval_loss": 1.5164753198623657,
      "eval_runtime": 156.2707,
      "eval_samples_per_second": 9.599,
      "eval_steps_per_second": 2.4,
      "step": 50
    },
    {
      "epoch": 0.025945945945945945,
      "grad_norm": 0.15546943247318268,
      "learning_rate": 6.474820143884892e-06,
      "loss": 1.5425,
      "step": 60
    },
    {
      "epoch": 0.034594594594594595,
      "grad_norm": 0.13446089625358582,
      "learning_rate": 8.633093525179856e-06,
      "loss": 1.5552,
      "step": 80
    },
    {
      "epoch": 0.043243243243243246,
      "grad_norm": 0.1574675589799881,
      "learning_rate": 1.079136690647482e-05,
      "loss": 1.4,
      "step": 100
    },
    {
      "epoch": 0.043243243243243246,
      "eval_loss": 1.511157751083374,
      "eval_runtime": 156.5135,
      "eval_samples_per_second": 9.584,
      "eval_steps_per_second": 2.396,
      "step": 100
    },
    {
      "epoch": 0.05189189189189189,
      "grad_norm": 0.1704365462064743,
      "learning_rate": 1.2949640287769784e-05,
      "loss": 1.3642,
      "step": 120
    },
    {
      "epoch": 0.06054054054054054,
      "grad_norm": 0.24820564687252045,
      "learning_rate": 1.4999998160049665e-05,
      "loss": 1.2921,
      "step": 140
    },
    {
      "epoch": 0.06486486486486487,
      "eval_loss": 1.472154140472412,
      "eval_runtime": 156.2248,
      "eval_samples_per_second": 9.602,
      "eval_steps_per_second": 2.4,
      "step": 150
    },
    {
      "epoch": 0.06918918918918919,
      "grad_norm": 0.19965757429599762,
      "learning_rate": 1.4999188596499784e-05,
      "loss": 1.3836,
      "step": 160
    },
    {
      "epoch": 0.07783783783783783,
      "grad_norm": 0.1957588493824005,
      "learning_rate": 1.4996907255939673e-05,
      "loss": 1.3328,
      "step": 180
    },
    {
      "epoch": 0.08648648648648649,
      "grad_norm": 0.2037293016910553,
      "learning_rate": 1.4993154586101044e-05,
      "loss": 1.4367,
      "step": 200
    },
    {
      "epoch": 0.08648648648648649,
      "eval_loss": 1.4281501770019531,
      "eval_runtime": 156.5874,
      "eval_samples_per_second": 9.579,
      "eval_steps_per_second": 2.395,
      "step": 200
    },
    {
      "epoch": 0.09513513513513513,
      "grad_norm": 0.2824411988258362,
      "learning_rate": 1.4987931323476007e-05,
      "loss": 1.4272,
      "step": 220
    },
    {
      "epoch": 0.10378378378378378,
      "grad_norm": 0.30133140087127686,
      "learning_rate": 1.4981238493172506e-05,
      "loss": 1.2994,
      "step": 240
    },
    {
      "epoch": 0.10810810810810811,
      "eval_loss": 1.3918442726135254,
      "eval_runtime": 156.3152,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 250
    },
    {
      "epoch": 0.11243243243243244,
      "grad_norm": 0.22853460907936096,
      "learning_rate": 1.4973077408713158e-05,
      "loss": 1.3668,
      "step": 260
    },
    {
      "epoch": 0.12108108108108108,
      "grad_norm": 0.25242647528648376,
      "learning_rate": 1.4963449671777435e-05,
      "loss": 1.3376,
      "step": 280
    },
    {
      "epoch": 0.12972972972972974,
      "grad_norm": 0.2533341348171234,
      "learning_rate": 1.4952357171887349e-05,
      "loss": 1.3953,
      "step": 300
    },
    {
      "epoch": 0.12972972972972974,
      "eval_loss": 1.3653450012207031,
      "eval_runtime": 156.5177,
      "eval_samples_per_second": 9.584,
      "eval_steps_per_second": 2.396,
      "step": 300
    },
    {
      "epoch": 0.13837837837837838,
      "grad_norm": 0.4832008481025696,
      "learning_rate": 1.4939802086036593e-05,
      "loss": 1.4049,
      "step": 320
    },
    {
      "epoch": 0.14702702702702702,
      "grad_norm": 0.27169403433799744,
      "learning_rate": 1.4925786878263313e-05,
      "loss": 1.4256,
      "step": 340
    },
    {
      "epoch": 0.15135135135135136,
      "eval_loss": 1.3452147245407104,
      "eval_runtime": 156.1759,
      "eval_samples_per_second": 9.605,
      "eval_steps_per_second": 2.401,
      "step": 350
    },
    {
      "epoch": 0.15567567567567567,
      "grad_norm": 0.42071032524108887,
      "learning_rate": 1.4910314299166496e-05,
      "loss": 1.3099,
      "step": 360
    },
    {
      "epoch": 0.1643243243243243,
      "grad_norm": 0.37401214241981506,
      "learning_rate": 1.4893387385366164e-05,
      "loss": 1.3174,
      "step": 380
    },
    {
      "epoch": 0.17297297297297298,
      "grad_norm": 0.36374762654304504,
      "learning_rate": 1.4875009458907393e-05,
      "loss": 1.325,
      "step": 400
    },
    {
      "epoch": 0.17297297297297298,
      "eval_loss": 1.3278440237045288,
      "eval_runtime": 155.9619,
      "eval_samples_per_second": 9.618,
      "eval_steps_per_second": 2.404,
      "step": 400
    },
    {
      "epoch": 0.18162162162162163,
      "grad_norm": 0.44585686922073364,
      "learning_rate": 1.4855184126608352e-05,
      "loss": 1.2749,
      "step": 420
    },
    {
      "epoch": 0.19027027027027027,
      "grad_norm": 0.40485116839408875,
      "learning_rate": 1.4833915279352421e-05,
      "loss": 1.1299,
      "step": 440
    },
    {
      "epoch": 0.1945945945945946,
      "eval_loss": 1.3141183853149414,
      "eval_runtime": 156.3105,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 450
    },
    {
      "epoch": 0.1989189189189189,
      "grad_norm": 0.4696143865585327,
      "learning_rate": 1.4811207091324587e-05,
      "loss": 1.1933,
      "step": 460
    },
    {
      "epoch": 0.20756756756756756,
      "grad_norm": 0.6063587069511414,
      "learning_rate": 1.478706401919221e-05,
      "loss": 1.2983,
      "step": 480
    },
    {
      "epoch": 0.21621621621621623,
      "grad_norm": 0.4177221953868866,
      "learning_rate": 1.4761490801230385e-05,
      "loss": 1.2339,
      "step": 500
    },
    {
      "epoch": 0.21621621621621623,
      "eval_loss": 1.302538275718689,
      "eval_runtime": 156.5384,
      "eval_samples_per_second": 9.582,
      "eval_steps_per_second": 2.396,
      "step": 500
    },
    {
      "epoch": 0.22486486486486487,
      "grad_norm": 0.4580596089363098,
      "learning_rate": 1.4734492456391998e-05,
      "loss": 1.3529,
      "step": 520
    },
    {
      "epoch": 0.23351351351351352,
      "grad_norm": 0.4583284556865692,
      "learning_rate": 1.470607428332273e-05,
      "loss": 1.3169,
      "step": 540
    },
    {
      "epoch": 0.23783783783783785,
      "eval_loss": 1.2932534217834473,
      "eval_runtime": 156.5697,
      "eval_samples_per_second": 9.58,
      "eval_steps_per_second": 2.395,
      "step": 550
    },
    {
      "epoch": 0.24216216216216216,
      "grad_norm": 0.4312450885772705,
      "learning_rate": 1.4676241859321143e-05,
      "loss": 1.1324,
      "step": 560
    },
    {
      "epoch": 0.2508108108108108,
      "grad_norm": 0.5200898051261902,
      "learning_rate": 1.4645001039244101e-05,
      "loss": 1.2754,
      "step": 580
    },
    {
      "epoch": 0.2594594594594595,
      "grad_norm": 0.5838780403137207,
      "learning_rate": 1.4612357954357695e-05,
      "loss": 1.2903,
      "step": 600
    },
    {
      "epoch": 0.2594594594594595,
      "eval_loss": 1.284968376159668,
      "eval_runtime": 156.4193,
      "eval_samples_per_second": 9.59,
      "eval_steps_per_second": 2.397,
      "step": 600
    },
    {
      "epoch": 0.2681081081081081,
      "grad_norm": 0.3660845458507538,
      "learning_rate": 1.4578319011133939e-05,
      "loss": 1.1267,
      "step": 620
    },
    {
      "epoch": 0.27675675675675676,
      "grad_norm": 0.4065239429473877,
      "learning_rate": 1.454289088999345e-05,
      "loss": 1.2034,
      "step": 640
    },
    {
      "epoch": 0.2810810810810811,
      "eval_loss": 1.2779653072357178,
      "eval_runtime": 156.5245,
      "eval_samples_per_second": 9.583,
      "eval_steps_per_second": 2.396,
      "step": 650
    },
    {
      "epoch": 0.28540540540540543,
      "grad_norm": 0.4448115825653076,
      "learning_rate": 1.4506080543994352e-05,
      "loss": 1.1839,
      "step": 660
    },
    {
      "epoch": 0.29405405405405405,
      "grad_norm": 0.741157054901123,
      "learning_rate": 1.4467895197467692e-05,
      "loss": 1.2143,
      "step": 680
    },
    {
      "epoch": 0.3027027027027027,
      "grad_norm": 0.45047101378440857,
      "learning_rate": 1.4428342344599592e-05,
      "loss": 1.2868,
      "step": 700
    },
    {
      "epoch": 0.3027027027027027,
      "eval_loss": 1.2726409435272217,
      "eval_runtime": 156.3492,
      "eval_samples_per_second": 9.594,
      "eval_steps_per_second": 2.398,
      "step": 700
    },
    {
      "epoch": 0.31135135135135134,
      "grad_norm": 0.4149288535118103,
      "learning_rate": 1.4387429747960472e-05,
      "loss": 1.1808,
      "step": 720
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.5024465322494507,
      "learning_rate": 1.4345165436981567e-05,
      "loss": 1.1616,
      "step": 740
    },
    {
      "epoch": 0.32432432432432434,
      "eval_loss": 1.2668416500091553,
      "eval_runtime": 156.232,
      "eval_samples_per_second": 9.601,
      "eval_steps_per_second": 2.4,
      "step": 750
    },
    {
      "epoch": 0.3286486486486486,
      "grad_norm": 0.5940514802932739,
      "learning_rate": 1.430155770637909e-05,
      "loss": 1.2955,
      "step": 760
    },
    {
      "epoch": 0.3372972972972973,
      "grad_norm": 0.4798080623149872,
      "learning_rate": 1.4256615114526333e-05,
      "loss": 1.1858,
      "step": 780
    },
    {
      "epoch": 0.34594594594594597,
      "grad_norm": 0.3747268319129944,
      "learning_rate": 1.4210346481774007e-05,
      "loss": 1.2592,
      "step": 800
    },
    {
      "epoch": 0.34594594594594597,
      "eval_loss": 1.261027216911316,
      "eval_runtime": 156.3223,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 800
    },
    {
      "epoch": 0.3545945945945946,
      "grad_norm": 0.6627951860427856,
      "learning_rate": 1.4162760888719174e-05,
      "loss": 1.1585,
      "step": 820
    },
    {
      "epoch": 0.36324324324324325,
      "grad_norm": 0.6270909309387207,
      "learning_rate": 1.4113867674423123e-05,
      "loss": 1.2351,
      "step": 840
    },
    {
      "epoch": 0.3675675675675676,
      "eval_loss": 1.2562013864517212,
      "eval_runtime": 156.3737,
      "eval_samples_per_second": 9.592,
      "eval_steps_per_second": 2.398,
      "step": 850
    },
    {
      "epoch": 0.37189189189189187,
      "grad_norm": 0.5605448484420776,
      "learning_rate": 1.4063676434578483e-05,
      "loss": 1.2151,
      "step": 860
    },
    {
      "epoch": 0.38054054054054054,
      "grad_norm": 0.4705483317375183,
      "learning_rate": 1.4012197019626004e-05,
      "loss": 1.1309,
      "step": 880
    },
    {
      "epoch": 0.3891891891891892,
      "grad_norm": 0.5667669773101807,
      "learning_rate": 1.3959439532821314e-05,
      "loss": 1.2531,
      "step": 900
    },
    {
      "epoch": 0.3891891891891892,
      "eval_loss": 1.2514050006866455,
      "eval_runtime": 156.3146,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 900
    },
    {
      "epoch": 0.3978378378378378,
      "grad_norm": 0.6878731846809387,
      "learning_rate": 1.3905414328252085e-05,
      "loss": 1.2473,
      "step": 920
    },
    {
      "epoch": 0.4064864864864865,
      "grad_norm": 0.799188494682312,
      "learning_rate": 1.3850132008805958e-05,
      "loss": 1.1098,
      "step": 940
    },
    {
      "epoch": 0.41081081081081083,
      "eval_loss": 1.2472807168960571,
      "eval_runtime": 156.3138,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 950
    },
    {
      "epoch": 0.4151351351351351,
      "grad_norm": 0.5942118763923645,
      "learning_rate": 1.379360342408963e-05,
      "loss": 1.1993,
      "step": 960
    },
    {
      "epoch": 0.4237837837837838,
      "grad_norm": 0.44030797481536865,
      "learning_rate": 1.3735839668299537e-05,
      "loss": 1.2577,
      "step": 980
    },
    {
      "epoch": 0.43243243243243246,
      "grad_norm": 0.5364145040512085,
      "learning_rate": 1.3676852078044528e-05,
      "loss": 1.209,
      "step": 1000
    },
    {
      "epoch": 0.43243243243243246,
      "eval_loss": 1.2433239221572876,
      "eval_runtime": 156.3683,
      "eval_samples_per_second": 9.593,
      "eval_steps_per_second": 2.398,
      "step": 1000
    },
    {
      "epoch": 0.4410810810810811,
      "grad_norm": 0.7929672598838806,
      "learning_rate": 1.3616652230120962e-05,
      "loss": 1.2357,
      "step": 1020
    },
    {
      "epoch": 0.44972972972972974,
      "grad_norm": 0.7581718564033508,
      "learning_rate": 1.3555251939240659e-05,
      "loss": 1.2092,
      "step": 1040
    },
    {
      "epoch": 0.4540540540540541,
      "eval_loss": 1.239459753036499,
      "eval_runtime": 156.2938,
      "eval_samples_per_second": 9.597,
      "eval_steps_per_second": 2.399,
      "step": 1050
    },
    {
      "epoch": 0.45837837837837836,
      "grad_norm": 0.5872713327407837,
      "learning_rate": 1.3492663255712187e-05,
      "loss": 1.2605,
      "step": 1060
    },
    {
      "epoch": 0.46702702702702703,
      "grad_norm": 0.5056009292602539,
      "learning_rate": 1.3428898463075862e-05,
      "loss": 1.2838,
      "step": 1080
    },
    {
      "epoch": 0.4756756756756757,
      "grad_norm": 0.5987468957901001,
      "learning_rate": 1.3363970075693032e-05,
      "loss": 1.2619,
      "step": 1100
    },
    {
      "epoch": 0.4756756756756757,
      "eval_loss": 1.2360332012176514,
      "eval_runtime": 156.4955,
      "eval_samples_per_second": 9.585,
      "eval_steps_per_second": 2.396,
      "step": 1100
    },
    {
      "epoch": 0.4843243243243243,
      "grad_norm": 0.6014208197593689,
      "learning_rate": 1.3297890836290004e-05,
      "loss": 1.2276,
      "step": 1120
    },
    {
      "epoch": 0.492972972972973,
      "grad_norm": 0.7187291383743286,
      "learning_rate": 1.3230673713457208e-05,
      "loss": 1.1775,
      "step": 1140
    },
    {
      "epoch": 0.4972972972972973,
      "eval_loss": 1.2334262132644653,
      "eval_runtime": 156.3371,
      "eval_samples_per_second": 9.595,
      "eval_steps_per_second": 2.399,
      "step": 1150
    },
    {
      "epoch": 0.5016216216216216,
      "grad_norm": 0.6566280722618103,
      "learning_rate": 1.3162331899103977e-05,
      "loss": 1.2047,
      "step": 1160
    },
    {
      "epoch": 0.5102702702702703,
      "grad_norm": 0.8335544466972351,
      "learning_rate": 1.3092878805869546e-05,
      "loss": 1.1767,
      "step": 1180
    },
    {
      "epoch": 0.518918918918919,
      "grad_norm": 0.6835708022117615,
      "learning_rate": 1.3022328064490712e-05,
      "loss": 1.1629,
      "step": 1200
    },
    {
      "epoch": 0.518918918918919,
      "eval_loss": 1.2302730083465576,
      "eval_runtime": 156.5356,
      "eval_samples_per_second": 9.582,
      "eval_steps_per_second": 2.396,
      "step": 1200
    },
    {
      "epoch": 0.5275675675675676,
      "grad_norm": 0.5832236409187317,
      "learning_rate": 1.2950693521126676e-05,
      "loss": 1.1748,
      "step": 1220
    },
    {
      "epoch": 0.5362162162162162,
      "grad_norm": 0.6758794188499451,
      "learning_rate": 1.2877989234641634e-05,
      "loss": 1.1371,
      "step": 1240
    },
    {
      "epoch": 0.5405405405405406,
      "eval_loss": 1.2266525030136108,
      "eval_runtime": 156.281,
      "eval_samples_per_second": 9.598,
      "eval_steps_per_second": 2.4,
      "step": 1250
    },
    {
      "epoch": 0.5448648648648649,
      "grad_norm": 0.5459420680999756,
      "learning_rate": 1.2804229473845609e-05,
      "loss": 1.2151,
      "step": 1260
    },
    {
      "epoch": 0.5535135135135135,
      "grad_norm": 0.5503604412078857,
      "learning_rate": 1.2729428714694075e-05,
      "loss": 1.1291,
      "step": 1280
    },
    {
      "epoch": 0.5621621621621622,
      "grad_norm": 0.6145665645599365,
      "learning_rate": 1.265360163744694e-05,
      "loss": 1.1212,
      "step": 1300
    },
    {
      "epoch": 0.5621621621621622,
      "eval_loss": 1.2243192195892334,
      "eval_runtime": 156.3088,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 1300
    },
    {
      "epoch": 0.5708108108108109,
      "grad_norm": 0.8270878791809082,
      "learning_rate": 1.257676312378741e-05,
      "loss": 1.0307,
      "step": 1320
    },
    {
      "epoch": 0.5794594594594594,
      "grad_norm": 0.6336683034896851,
      "learning_rate": 1.2498928253901358e-05,
      "loss": 1.143,
      "step": 1340
    },
    {
      "epoch": 0.5837837837837838,
      "eval_loss": 1.2215410470962524,
      "eval_runtime": 156.2606,
      "eval_samples_per_second": 9.599,
      "eval_steps_per_second": 2.4,
      "step": 1350
    },
    {
      "epoch": 0.5881081081081081,
      "grad_norm": 0.5482487678527832,
      "learning_rate": 1.24201123035177e-05,
      "loss": 1.3624,
      "step": 1360
    },
    {
      "epoch": 0.5967567567567568,
      "grad_norm": 0.7318127751350403,
      "learning_rate": 1.2340330740910413e-05,
      "loss": 1.1077,
      "step": 1380
    },
    {
      "epoch": 0.6054054054054054,
      "grad_norm": 0.5615875720977783,
      "learning_rate": 1.2259599223862757e-05,
      "loss": 1.232,
      "step": 1400
    },
    {
      "epoch": 0.6054054054054054,
      "eval_loss": 1.2196455001831055,
      "eval_runtime": 156.1978,
      "eval_samples_per_second": 9.603,
      "eval_steps_per_second": 2.401,
      "step": 1400
    },
    {
      "epoch": 0.614054054054054,
      "grad_norm": 0.6447334885597229,
      "learning_rate": 1.2177933596594313e-05,
      "loss": 1.1058,
      "step": 1420
    },
    {
      "epoch": 0.6227027027027027,
      "grad_norm": 0.8408645391464233,
      "learning_rate": 1.2095349886651426e-05,
      "loss": 1.302,
      "step": 1440
    },
    {
      "epoch": 0.6270270270270271,
      "eval_loss": 1.2167232036590576,
      "eval_runtime": 156.2131,
      "eval_samples_per_second": 9.602,
      "eval_steps_per_second": 2.401,
      "step": 1450
    },
    {
      "epoch": 0.6313513513513513,
      "grad_norm": 0.4599859416484833,
      "learning_rate": 1.2011864301761648e-05,
      "loss": 1.1504,
      "step": 1460
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.9404271841049194,
      "learning_rate": 1.1927493226652876e-05,
      "loss": 1.2024,
      "step": 1480
    },
    {
      "epoch": 0.6486486486486487,
      "grad_norm": 0.5044443011283875,
      "learning_rate": 1.1842253219837676e-05,
      "loss": 1.213,
      "step": 1500
    },
    {
      "epoch": 0.6486486486486487,
      "eval_loss": 1.2149133682250977,
      "eval_runtime": 156.3249,
      "eval_samples_per_second": 9.595,
      "eval_steps_per_second": 2.399,
      "step": 1500
    },
    {
      "epoch": 0.6572972972972972,
      "grad_norm": 0.7133287191390991,
      "learning_rate": 1.1756161010363576e-05,
      "loss": 1.1682,
      "step": 1520
    },
    {
      "epoch": 0.6659459459459459,
      "grad_norm": 0.7438573241233826,
      "learning_rate": 1.1669233494529819e-05,
      "loss": 1.0579,
      "step": 1540
    },
    {
      "epoch": 0.6702702702702703,
      "eval_loss": 1.2130820751190186,
      "eval_runtime": 156.4016,
      "eval_samples_per_second": 9.591,
      "eval_steps_per_second": 2.398,
      "step": 1550
    },
    {
      "epoch": 0.6745945945945946,
      "grad_norm": 0.8739683032035828,
      "learning_rate": 1.158148773257135e-05,
      "loss": 1.0753,
      "step": 1560
    },
    {
      "epoch": 0.6832432432432433,
      "grad_norm": 0.7708102464675903,
      "learning_rate": 1.1492940945310608e-05,
      "loss": 1.2175,
      "step": 1580
    },
    {
      "epoch": 0.6918918918918919,
      "grad_norm": 0.549228847026825,
      "learning_rate": 1.1403610510777778e-05,
      "loss": 1.0838,
      "step": 1600
    },
    {
      "epoch": 0.6918918918918919,
      "eval_loss": 1.2114548683166504,
      "eval_runtime": 156.4625,
      "eval_samples_per_second": 9.587,
      "eval_steps_per_second": 2.397,
      "step": 1600
    },
    {
      "epoch": 0.7005405405405405,
      "grad_norm": 0.9724882245063782,
      "learning_rate": 1.1313513960800224e-05,
      "loss": 1.0542,
      "step": 1620
    },
    {
      "epoch": 0.7091891891891892,
      "grad_norm": 0.8888576030731201,
      "learning_rate": 1.1222668977561725e-05,
      "loss": 1.2136,
      "step": 1640
    },
    {
      "epoch": 0.7135135135135136,
      "eval_loss": 1.2095521688461304,
      "eval_runtime": 156.4366,
      "eval_samples_per_second": 9.589,
      "eval_steps_per_second": 2.397,
      "step": 1650
    },
    {
      "epoch": 0.7178378378378378,
      "grad_norm": 1.1320910453796387,
      "learning_rate": 1.1131093390132205e-05,
      "loss": 1.1844,
      "step": 1660
    },
    {
      "epoch": 0.7264864864864865,
      "grad_norm": 0.8856949210166931,
      "learning_rate": 1.1038805170968615e-05,
      "loss": 1.1142,
      "step": 1680
    },
    {
      "epoch": 0.7351351351351352,
      "grad_norm": 1.0155227184295654,
      "learning_rate": 1.0945822432387722e-05,
      "loss": 1.1138,
      "step": 1700
    },
    {
      "epoch": 0.7351351351351352,
      "eval_loss": 1.207614779472351,
      "eval_runtime": 156.4751,
      "eval_samples_per_second": 9.586,
      "eval_steps_per_second": 2.397,
      "step": 1700
    },
    {
      "epoch": 0.7437837837837837,
      "grad_norm": 0.8122560977935791,
      "learning_rate": 1.0852163423011382e-05,
      "loss": 1.1304,
      "step": 1720
    },
    {
      "epoch": 0.7524324324324324,
      "grad_norm": 0.8628650903701782,
      "learning_rate": 1.0757846524185135e-05,
      "loss": 1.1426,
      "step": 1740
    },
    {
      "epoch": 0.7567567567567568,
      "eval_loss": 1.2060997486114502,
      "eval_runtime": 156.2117,
      "eval_samples_per_second": 9.602,
      "eval_steps_per_second": 2.401,
      "step": 1750
    },
    {
      "epoch": 0.7610810810810811,
      "grad_norm": 1.243729591369629,
      "learning_rate": 1.0662890246370682e-05,
      "loss": 1.0881,
      "step": 1760
    },
    {
      "epoch": 0.7697297297297298,
      "grad_norm": 0.8714478015899658,
      "learning_rate": 1.0567313225513085e-05,
      "loss": 1.2214,
      "step": 1780
    },
    {
      "epoch": 0.7783783783783784,
      "grad_norm": 0.44382530450820923,
      "learning_rate": 1.0471134219383299e-05,
      "loss": 1.2328,
      "step": 1800
    },
    {
      "epoch": 0.7783783783783784,
      "eval_loss": 1.2047566175460815,
      "eval_runtime": 156.372,
      "eval_samples_per_second": 9.593,
      "eval_steps_per_second": 2.398,
      "step": 1800
    },
    {
      "epoch": 0.787027027027027,
      "grad_norm": 1.0515871047973633,
      "learning_rate": 1.0374372103896792e-05,
      "loss": 1.1519,
      "step": 1820
    },
    {
      "epoch": 0.7956756756756757,
      "grad_norm": 0.8955100178718567,
      "learning_rate": 1.027704586940902e-05,
      "loss": 1.0786,
      "step": 1840
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.2030872106552124,
      "eval_runtime": 156.2817,
      "eval_samples_per_second": 9.598,
      "eval_steps_per_second": 2.4,
      "step": 1850
    },
    {
      "epoch": 0.8043243243243243,
      "grad_norm": 0.6575765609741211,
      "learning_rate": 1.0179174616988392e-05,
      "loss": 1.2385,
      "step": 1860
    },
    {
      "epoch": 0.812972972972973,
      "grad_norm": 0.5790345072746277,
      "learning_rate": 1.008077755466754e-05,
      "loss": 1.135,
      "step": 1880
    },
    {
      "epoch": 0.8216216216216217,
      "grad_norm": 0.7308002710342407,
      "learning_rate": 9.981873993673584e-06,
      "loss": 1.2052,
      "step": 1900
    },
    {
      "epoch": 0.8216216216216217,
      "eval_loss": 1.2018324136734009,
      "eval_runtime": 156.3907,
      "eval_samples_per_second": 9.591,
      "eval_steps_per_second": 2.398,
      "step": 1900
    },
    {
      "epoch": 0.8302702702702702,
      "grad_norm": 0.47168049216270447,
      "learning_rate": 9.882483344638144e-06,
      "loss": 1.1291,
      "step": 1920
    },
    {
      "epoch": 0.8389189189189189,
      "grad_norm": 1.0765182971954346,
      "learning_rate": 9.78262511378785e-06,
      "loss": 1.2196,
      "step": 1940
    },
    {
      "epoch": 0.8432432432432433,
      "eval_loss": 1.200095534324646,
      "eval_runtime": 156.3628,
      "eval_samples_per_second": 9.593,
      "eval_steps_per_second": 2.398,
      "step": 1950
    },
    {
      "epoch": 0.8475675675675676,
      "grad_norm": 0.9777532815933228,
      "learning_rate": 9.682318899116075e-06,
      "loss": 1.2965,
      "step": 1960
    },
    {
      "epoch": 0.8562162162162162,
      "grad_norm": 0.6992659568786621,
      "learning_rate": 9.58158438653668e-06,
      "loss": 1.2705,
      "step": 1980
    },
    {
      "epoch": 0.8648648648648649,
      "grad_norm": 0.7154545783996582,
      "learning_rate": 9.480441346020491e-06,
      "loss": 1.1538,
      "step": 2000
    },
    {
      "epoch": 0.8648648648648649,
      "eval_loss": 1.1992037296295166,
      "eval_runtime": 156.5695,
      "eval_samples_per_second": 9.58,
      "eval_steps_per_second": 2.395,
      "step": 2000
    },
    {
      "epoch": 0.8735135135135135,
      "grad_norm": 0.9878578782081604,
      "learning_rate": 9.378909627715276e-06,
      "loss": 1.1575,
      "step": 2020
    },
    {
      "epoch": 0.8821621621621621,
      "grad_norm": 0.7160919308662415,
      "learning_rate": 9.277009158049991e-06,
      "loss": 1.2407,
      "step": 2040
    },
    {
      "epoch": 0.8864864864864865,
      "eval_loss": 1.1983826160430908,
      "eval_runtime": 156.5345,
      "eval_samples_per_second": 9.583,
      "eval_steps_per_second": 2.396,
      "step": 2050
    },
    {
      "epoch": 0.8908108108108108,
      "grad_norm": 0.8028165102005005,
      "learning_rate": 9.174759935824059e-06,
      "loss": 1.0845,
      "step": 2060
    },
    {
      "epoch": 0.8994594594594595,
      "grad_norm": 0.5290041565895081,
      "learning_rate": 9.072182028282447e-06,
      "loss": 1.1363,
      "step": 2080
    },
    {
      "epoch": 0.9081081081081082,
      "grad_norm": 0.8659726977348328,
      "learning_rate": 8.96929556717729e-06,
      "loss": 1.1844,
      "step": 2100
    },
    {
      "epoch": 0.9081081081081082,
      "eval_loss": 1.1970139741897583,
      "eval_runtime": 156.645,
      "eval_samples_per_second": 9.576,
      "eval_steps_per_second": 2.394,
      "step": 2100
    },
    {
      "epoch": 0.9167567567567567,
      "grad_norm": 0.7906278371810913,
      "learning_rate": 8.866120744816888e-06,
      "loss": 1.1765,
      "step": 2120
    },
    {
      "epoch": 0.9254054054054054,
      "grad_norm": 0.7997767925262451,
      "learning_rate": 8.762677810102787e-06,
      "loss": 1.0805,
      "step": 2140
    },
    {
      "epoch": 0.9297297297297298,
      "eval_loss": 1.195316195487976,
      "eval_runtime": 156.3354,
      "eval_samples_per_second": 9.595,
      "eval_steps_per_second": 2.399,
      "step": 2150
    },
    {
      "epoch": 0.9340540540540541,
      "grad_norm": 0.5412635803222656,
      "learning_rate": 8.658987064555783e-06,
      "loss": 1.0922,
      "step": 2160
    },
    {
      "epoch": 0.9427027027027027,
      "grad_norm": 0.8357748985290527,
      "learning_rate": 8.555068858331553e-06,
      "loss": 1.2273,
      "step": 2180
    },
    {
      "epoch": 0.9513513513513514,
      "grad_norm": 0.8305333852767944,
      "learning_rate": 8.450943586226805e-06,
      "loss": 1.2381,
      "step": 2200
    },
    {
      "epoch": 0.9513513513513514,
      "eval_loss": 1.1945637464523315,
      "eval_runtime": 157.8154,
      "eval_samples_per_second": 9.505,
      "eval_steps_per_second": 2.376,
      "step": 2200
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5756638646125793,
      "learning_rate": 8.346631683676616e-06,
      "loss": 1.0844,
      "step": 2220
    },
    {
      "epoch": 0.9686486486486486,
      "grad_norm": 0.8370802998542786,
      "learning_rate": 8.242153622743802e-06,
      "loss": 1.1966,
      "step": 2240
    },
    {
      "epoch": 0.972972972972973,
      "eval_loss": 1.193472981452942,
      "eval_runtime": 156.5861,
      "eval_samples_per_second": 9.579,
      "eval_steps_per_second": 2.395,
      "step": 2250
    },
    {
      "epoch": 0.9772972972972973,
      "grad_norm": 0.654332160949707,
      "learning_rate": 8.137529908101128e-06,
      "loss": 1.1768,
      "step": 2260
    },
    {
      "epoch": 0.985945945945946,
      "grad_norm": 0.8234332799911499,
      "learning_rate": 8.03278107300709e-06,
      "loss": 1.1339,
      "step": 2280
    },
    {
      "epoch": 0.9945945945945946,
      "grad_norm": 0.618282675743103,
      "learning_rate": 7.927927675276087e-06,
      "loss": 1.1226,
      "step": 2300
    },
    {
      "epoch": 0.9945945945945946,
      "eval_loss": 1.1929038763046265,
      "eval_runtime": 156.4657,
      "eval_samples_per_second": 9.587,
      "eval_steps_per_second": 2.397,
      "step": 2300
    },
    {
      "epoch": 1.0032432432432432,
      "grad_norm": 0.7228516340255737,
      "learning_rate": 7.822990293243793e-06,
      "loss": 1.1974,
      "step": 2320
    },
    {
      "epoch": 1.011891891891892,
      "grad_norm": 0.6766774654388428,
      "learning_rate": 7.71798952172849e-06,
      "loss": 1.01,
      "step": 2340
    },
    {
      "epoch": 1.0162162162162163,
      "eval_loss": 1.1917825937271118,
      "eval_runtime": 156.3134,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 2350
    },
    {
      "epoch": 1.0205405405405406,
      "grad_norm": 0.8917098045349121,
      "learning_rate": 7.612945967989164e-06,
      "loss": 1.0233,
      "step": 2360
    },
    {
      "epoch": 1.0291891891891891,
      "grad_norm": 0.7241379022598267,
      "learning_rate": 7.5078802476811515e-06,
      "loss": 1.2105,
      "step": 2380
    },
    {
      "epoch": 1.037837837837838,
      "grad_norm": 0.6121885776519775,
      "learning_rate": 7.402812980810174e-06,
      "loss": 1.1697,
      "step": 2400
    },
    {
      "epoch": 1.037837837837838,
      "eval_loss": 1.1907098293304443,
      "eval_runtime": 156.2571,
      "eval_samples_per_second": 9.6,
      "eval_steps_per_second": 2.4,
      "step": 2400
    },
    {
      "epoch": 1.0464864864864865,
      "grad_norm": 0.56922847032547,
      "learning_rate": 7.297764787685467e-06,
      "loss": 1.2328,
      "step": 2420
    },
    {
      "epoch": 1.0551351351351352,
      "grad_norm": 0.8866540789604187,
      "learning_rate": 7.192756284872893e-06,
      "loss": 1.1378,
      "step": 2440
    },
    {
      "epoch": 1.0594594594594595,
      "eval_loss": 1.190106987953186,
      "eval_runtime": 156.2921,
      "eval_samples_per_second": 9.597,
      "eval_steps_per_second": 2.399,
      "step": 2450
    },
    {
      "epoch": 1.0637837837837838,
      "grad_norm": 0.9187541604042053,
      "learning_rate": 7.087808081148768e-06,
      "loss": 1.027,
      "step": 2460
    },
    {
      "epoch": 1.0724324324324324,
      "grad_norm": 0.4501543939113617,
      "learning_rate": 6.982940773455215e-06,
      "loss": 1.2174,
      "step": 2480
    },
    {
      "epoch": 1.0810810810810811,
      "grad_norm": 0.670372486114502,
      "learning_rate": 6.878174942857852e-06,
      "loss": 0.9583,
      "step": 2500
    },
    {
      "epoch": 1.0810810810810811,
      "eval_loss": 1.1890244483947754,
      "eval_runtime": 156.3208,
      "eval_samples_per_second": 9.596,
      "eval_steps_per_second": 2.399,
      "step": 2500
    },
    {
      "epoch": 1.0897297297297297,
      "grad_norm": 0.5464470386505127,
      "learning_rate": 6.7735311505065885e-06,
      "loss": 1.117,
      "step": 2520
    },
    {
      "epoch": 1.0983783783783783,
      "grad_norm": 0.5962574481964111,
      "learning_rate": 6.669029933600332e-06,
      "loss": 1.1578,
      "step": 2540
    },
    {
      "epoch": 1.1027027027027028,
      "eval_loss": 1.1883645057678223,
      "eval_runtime": 156.3484,
      "eval_samples_per_second": 9.594,
      "eval_steps_per_second": 2.398,
      "step": 2550
    },
    {
      "epoch": 1.107027027027027,
      "grad_norm": 0.7612497210502625,
      "learning_rate": 6.564691801356392e-06,
      "loss": 1.1369,
      "step": 2560
    },
    {
      "epoch": 1.1156756756756756,
      "grad_norm": 1.0087133646011353,
      "learning_rate": 6.460537230985388e-06,
      "loss": 1.1793,
      "step": 2580
    },
    {
      "epoch": 1.1243243243243244,
      "grad_norm": 0.6867474913597107,
      "learning_rate": 6.356586663672409e-06,
      "loss": 1.1688,
      "step": 2600
    },
    {
      "epoch": 1.1243243243243244,
      "eval_loss": 1.1876789331436157,
      "eval_runtime": 156.4562,
      "eval_samples_per_second": 9.587,
      "eval_steps_per_second": 2.397,
      "step": 2600
    },
    {
      "epoch": 1.132972972972973,
      "grad_norm": 0.8074759840965271,
      "learning_rate": 6.252860500565295e-06,
      "loss": 1.1414,
      "step": 2620
    },
    {
      "epoch": 1.1416216216216215,
      "grad_norm": 1.0738656520843506,
      "learning_rate": 6.1493790987707146e-06,
      "loss": 1.2174,
      "step": 2640
    },
    {
      "epoch": 1.145945945945946,
      "eval_loss": 1.1870859861373901,
      "eval_runtime": 156.2871,
      "eval_samples_per_second": 9.598,
      "eval_steps_per_second": 2.399,
      "step": 2650
    },
    {
      "epoch": 1.1502702702702703,
      "grad_norm": 0.7977942824363708,
      "learning_rate": 6.0461627673589336e-06,
      "loss": 1.2674,
      "step": 2660
    },
    {
      "epoch": 1.1589189189189189,
      "grad_norm": 0.5102550983428955,
      "learning_rate": 5.943231763377995e-06,
      "loss": 1.1778,
      "step": 2680
    },
    {
      "epoch": 1.1675675675675676,
      "grad_norm": 0.7279404997825623,
      "learning_rate": 5.840606287878094e-06,
      "loss": 1.1715,
      "step": 2700
    },
    {
      "epoch": 1.1675675675675676,
      "eval_loss": 1.1867084503173828,
      "eval_runtime": 156.3684,
      "eval_samples_per_second": 9.593,
      "eval_steps_per_second": 2.398,
      "step": 2700
    },
    {
      "epoch": 1.1762162162162162,
      "grad_norm": 1.0068808794021606,
      "learning_rate": 5.738306481946973e-06,
      "loss": 1.1962,
      "step": 2720
    },
    {
      "epoch": 1.1848648648648648,
      "grad_norm": 1.274601936340332,
      "learning_rate": 5.636352422757062e-06,
      "loss": 1.1244,
      "step": 2740
    },
    {
      "epoch": 1.1891891891891893,
      "eval_loss": 1.185834288597107,
      "eval_runtime": 156.5394,
      "eval_samples_per_second": 9.582,
      "eval_steps_per_second": 2.396,
      "step": 2750
    },
    {
      "epoch": 1.1935135135135135,
      "grad_norm": 0.7775521278381348,
      "learning_rate": 5.534764119625167e-06,
      "loss": 1.1735,
      "step": 2760
    },
    {
      "epoch": 1.202162162162162,
      "grad_norm": 0.7912615537643433,
      "learning_rate": 5.433561510085475e-06,
      "loss": 1.1918,
      "step": 2780
    },
    {
      "epoch": 1.2108108108108109,
      "grad_norm": 0.7740307450294495,
      "learning_rate": 5.3327644559766674e-06,
      "loss": 1.0983,
      "step": 2800
    },
    {
      "epoch": 1.2108108108108109,
      "eval_loss": 1.185600757598877,
      "eval_runtime": 156.2111,
      "eval_samples_per_second": 9.602,
      "eval_steps_per_second": 2.401,
      "step": 2800
    },
    {
      "epoch": 1.2194594594594594,
      "grad_norm": 0.7494848370552063,
      "learning_rate": 5.232392739543848e-06,
      "loss": 1.0842,
      "step": 2820
    },
    {
      "epoch": 1.228108108108108,
      "grad_norm": 0.936766505241394,
      "learning_rate": 5.132466059556139e-06,
      "loss": 1.0828,
      "step": 2840
    },
    {
      "epoch": 1.2324324324324325,
      "eval_loss": 1.184820294380188,
      "eval_runtime": 156.2295,
      "eval_samples_per_second": 9.601,
      "eval_steps_per_second": 2.4,
      "step": 2850
    }
  ],
  "logging_steps": 20,
  "max_steps": 4624,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 2,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.08620239192064e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}