{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9752701960065946,
  "eval_steps": 100,
  "global_step": 5391,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0001999998159397344,
      "loss": 1.4113,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019999926375961516,
      "loss": 0.8357,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019999834346167496,
      "loss": 0.7122,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001999970550493016,
      "loss": 0.6653,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000199995398527238,
      "loss": 0.6152,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019999337390158218,
      "loss": 0.5963,
      "step": 60
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019999098117978715,
      "loss": 0.5923,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019998822037066105,
      "loss": 0.5783,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019998509148436697,
      "loss": 0.5529,
      "step": 90
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.000199981594532423,
      "loss": 0.5382,
      "step": 100
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.5621132850646973,
      "eval_runtime": 398.8595,
      "eval_samples_per_second": 11.992,
      "eval_steps_per_second": 2.999,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019997772952770216,
      "loss": 0.5489,
      "step": 110
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019997349648443225,
      "loss": 0.5268,
      "step": 120
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019996889541819602,
      "loss": 0.525,
      "step": 130
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019996392634593092,
      "loss": 0.4926,
      "step": 140
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019995858928592916,
      "loss": 0.5181,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019995288425783754,
      "loss": 0.5225,
      "step": 160
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019994681128265743,
      "loss": 0.4993,
      "step": 170
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019994037038274467,
      "loss": 0.5387,
      "step": 180
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001999335615818096,
      "loss": 0.465,
      "step": 190
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019992638490491676,
      "loss": 0.4788,
      "step": 200
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.5201168060302734,
      "eval_runtime": 398.3332,
      "eval_samples_per_second": 12.008,
      "eval_steps_per_second": 3.003,
      "step": 200
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019991884037848497,
      "loss": 0.451,
      "step": 210
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019991092803028725,
      "loss": 0.4586,
      "step": 220
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019990264788945052,
      "loss": 0.4894,
      "step": 230
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019989399998645568,
      "loss": 0.4666,
      "step": 240
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019988498435313744,
      "loss": 0.4763,
      "step": 250
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019987560102268422,
      "loss": 0.4539,
      "step": 260
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019986585002963793,
      "loss": 0.5133,
      "step": 270
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019985573140989405,
      "loss": 0.4704,
      "step": 280
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019984524520070125,
      "loss": 0.4548,
      "step": 290
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019983439144066143,
      "loss": 0.4682,
      "step": 300
    },
    {
      "epoch": 0.11,
      "eval_loss": 0.497054785490036,
      "eval_runtime": 399.4057,
      "eval_samples_per_second": 11.975,
      "eval_steps_per_second": 2.994,
      "step": 300
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001998231701697295,
      "loss": 0.4867,
      "step": 310
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001998115814292133,
      "loss": 0.4486,
      "step": 320
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001997996252617733,
      "loss": 0.4472,
      "step": 330
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019978730171142268,
      "loss": 0.4821,
      "step": 340
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001997746108235269,
      "loss": 0.4475,
      "step": 350
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019976155264480377,
      "loss": 0.4485,
      "step": 360
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019974812722332308,
      "loss": 0.4547,
      "step": 370
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001997343346085066,
      "loss": 0.4279,
      "step": 380
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019972017485112774,
      "loss": 0.4748,
      "step": 390
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001997056480033115,
      "loss": 0.4778,
      "step": 400
    },
    {
      "epoch": 0.15,
      "eval_loss": 0.4730743169784546,
      "eval_runtime": 403.2392,
      "eval_samples_per_second": 11.861,
      "eval_steps_per_second": 2.966,
      "step": 400
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001996907541185342,
      "loss": 0.4393,
      "step": 410
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019967549325162324,
      "loss": 0.4562,
      "step": 420
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019965986545875708,
      "loss": 0.4389,
      "step": 430
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001996438707974648,
      "loss": 0.4282,
      "step": 440
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000199627509326626,
      "loss": 0.4254,
      "step": 450
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001996107811064706,
      "loss": 0.419,
      "step": 460
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019959368619857872,
      "loss": 0.4261,
      "step": 470
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001995762246658801,
      "loss": 0.4581,
      "step": 480
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019955839657265432,
      "loss": 0.4333,
      "step": 490
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019954020198453018,
      "loss": 0.4541,
      "step": 500
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.46874529123306274,
      "eval_runtime": 403.7251,
      "eval_samples_per_second": 11.847,
      "eval_steps_per_second": 2.962,
      "step": 500
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019952164096848578,
      "loss": 0.4482,
      "step": 510
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019950271359284795,
      "loss": 0.4475,
      "step": 520
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019948341992729227,
      "loss": 0.4339,
      "step": 530
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019946376004284272,
      "loss": 0.4527,
      "step": 540
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001994437340118713,
      "loss": 0.4009,
      "step": 550
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.000199423341908098,
      "loss": 0.4496,
      "step": 560
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001994025838065903,
      "loss": 0.4314,
      "step": 570
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001993814597837631,
      "loss": 0.4454,
      "step": 580
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019935996991737818,
      "loss": 0.4076,
      "step": 590
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001993381142865442,
      "loss": 0.4786,
      "step": 600
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.4500006139278412,
      "eval_runtime": 399.7496,
      "eval_samples_per_second": 11.965,
      "eval_steps_per_second": 2.992,
      "step": 600
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019931589297171628,
      "loss": 0.4386,
      "step": 610
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001992933060546955,
      "loss": 0.4805,
      "step": 620
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019927035361862904,
      "loss": 0.4135,
      "step": 630
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001992470357480095,
      "loss": 0.421,
      "step": 640
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019922335252867476,
      "loss": 0.4224,
      "step": 650
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019919930404780766,
      "loss": 0.4324,
      "step": 660
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001991748903939355,
      "loss": 0.4062,
      "step": 670
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019915011165692997,
      "loss": 0.4107,
      "step": 680
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019912496792800677,
      "loss": 0.3953,
      "step": 690
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019909945929972502,
      "loss": 0.3974,
      "step": 700
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.4456511437892914,
      "eval_runtime": 403.5923,
      "eval_samples_per_second": 11.851,
      "eval_steps_per_second": 2.963,
      "step": 700
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001990735858659873,
      "loss": 0.4006,
      "step": 710
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000199047347722039,
      "loss": 0.4258,
      "step": 720
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019902074496446815,
      "loss": 0.4083,
      "step": 730
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019899377769120487,
      "loss": 0.4097,
      "step": 740
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019896644600152135,
      "loss": 0.4262,
      "step": 750
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019893874999603103,
      "loss": 0.4204,
      "step": 760
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001989106897766887,
      "loss": 0.4233,
      "step": 770
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001988822654467897,
      "loss": 0.4107,
      "step": 780
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019885347711096993,
      "loss": 0.4013,
      "step": 790
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019882432487520506,
      "loss": 0.4142,
      "step": 800
    },
    {
      "epoch": 0.29,
      "eval_loss": 0.4460389018058777,
      "eval_runtime": 400.2256,
      "eval_samples_per_second": 11.951,
      "eval_steps_per_second": 2.988,
      "step": 800
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001987948088468105,
      "loss": 0.4324,
      "step": 810
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001987649291344408,
      "loss": 0.3975,
      "step": 820
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019873468584808934,
      "loss": 0.4109,
      "step": 830
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019870407909908786,
      "loss": 0.3987,
      "step": 840
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019867310900010605,
      "loss": 0.3885,
      "step": 850
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019864177566515122,
      "loss": 0.4122,
      "step": 860
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019861007920956786,
      "loss": 0.405,
      "step": 870
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019857801975003704,
      "loss": 0.404,
      "step": 880
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001985455974045763,
      "loss": 0.382,
      "step": 890
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001985128122925389,
      "loss": 0.4374,
      "step": 900
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.44203105568885803,
      "eval_runtime": 401.2967,
      "eval_samples_per_second": 11.919,
      "eval_steps_per_second": 2.98,
      "step": 900
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019847966453461358,
      "loss": 0.3859,
      "step": 910
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019844615425282405,
      "loss": 0.4187,
      "step": 920
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019841228157052853,
      "loss": 0.4078,
      "step": 930
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001983780466124193,
      "loss": 0.4218,
      "step": 940
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001983434495045223,
      "loss": 0.4109,
      "step": 950
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019830849037419656,
      "loss": 0.4249,
      "step": 960
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019827316935013388,
      "loss": 0.4073,
      "step": 970
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001982374865623581,
      "loss": 0.4336,
      "step": 980
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019820144214222497,
      "loss": 0.4056,
      "step": 990
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019816503622242137,
      "loss": 0.4008,
      "step": 1000
    },
    {
      "epoch": 0.37,
      "eval_loss": 0.4418930411338806,
      "eval_runtime": 404.9805,
      "eval_samples_per_second": 11.81,
      "eval_steps_per_second": 2.953,
      "step": 1000
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019812826893696495,
      "loss": 0.365,
      "step": 1010
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019809114042120367,
      "loss": 0.4006,
      "step": 1020
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001980536508118152,
      "loss": 0.3714,
      "step": 1030
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019801580024680652,
      "loss": 0.3945,
      "step": 1040
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019797758886551324,
      "loss": 0.4316,
      "step": 1050
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001979390168085994,
      "loss": 0.3958,
      "step": 1060
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019790008421805664,
      "loss": 0.419,
      "step": 1070
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019786079123720377,
      "loss": 0.4048,
      "step": 1080
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001978211380106864,
      "loss": 0.4002,
      "step": 1090
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001977811246844761,
      "loss": 0.3979,
      "step": 1100
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.43328720331192017,
      "eval_runtime": 398.734,
      "eval_samples_per_second": 11.995,
      "eval_steps_per_second": 2.999,
      "step": 1100
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019774075140587024,
      "loss": 0.4047,
      "step": 1110
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019770001832349106,
      "loss": 0.389,
      "step": 1120
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019765892558728542,
      "loss": 0.3628,
      "step": 1130
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001976174733485242,
      "loss": 0.3914,
      "step": 1140
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019757566175980146,
      "loss": 0.3885,
      "step": 1150
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019753349097503437,
      "loss": 0.4023,
      "step": 1160
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001974909611494622,
      "loss": 0.4358,
      "step": 1170
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019744807243964597,
      "loss": 0.3965,
      "step": 1180
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019740482500346779,
      "loss": 0.3762,
      "step": 1190
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001973612190001304,
      "loss": 0.4108,
      "step": 1200
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.4304308295249939,
      "eval_runtime": 397.6793,
      "eval_samples_per_second": 12.027,
      "eval_steps_per_second": 3.007,
      "step": 1200
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019731725459015643,
      "loss": 0.3838,
      "step": 1210
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019727293193538793,
      "loss": 0.3782,
      "step": 1220
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019722825119898566,
      "loss": 0.4034,
      "step": 1230
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019718321254542858,
      "loss": 0.391,
      "step": 1240
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001971378161405132,
      "loss": 0.4082,
      "step": 1250
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001970920621513531,
      "loss": 0.3772,
      "step": 1260
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019704595074637805,
      "loss": 0.3894,
      "step": 1270
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019699948209533355,
      "loss": 0.3882,
      "step": 1280
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019695265636928032,
      "loss": 0.3782,
      "step": 1290
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001969054737405934,
      "loss": 0.3578,
      "step": 1300
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.42550337314605713,
      "eval_runtime": 402.9262,
      "eval_samples_per_second": 11.871,
      "eval_steps_per_second": 2.968,
      "step": 1300
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019685793438296183,
      "loss": 0.3644,
      "step": 1310
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019681003847138765,
      "loss": 0.3702,
      "step": 1320
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019676178618218565,
      "loss": 0.4213,
      "step": 1330
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001967131776929823,
      "loss": 0.4126,
      "step": 1340
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019666421318271547,
      "loss": 0.3643,
      "step": 1350
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019661489283163362,
      "loss": 0.3854,
      "step": 1360
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019656521682129502,
      "loss": 0.3933,
      "step": 1370
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019651518533456733,
      "loss": 0.3841,
      "step": 1380
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019646479855562666,
      "loss": 0.3677,
      "step": 1390
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019641405666995715,
      "loss": 0.3895,
      "step": 1400
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.41961902379989624,
      "eval_runtime": 402.2386,
      "eval_samples_per_second": 11.891,
      "eval_steps_per_second": 2.973,
      "step": 1400
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019636295986435003,
      "loss": 0.3578,
      "step": 1410
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019631150832690318,
      "loss": 0.3878,
      "step": 1420
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019625970224702025,
      "loss": 0.3784,
      "step": 1430
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019620754181541008,
      "loss": 0.4042,
      "step": 1440
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001961550272240859,
      "loss": 0.361,
      "step": 1450
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019610215866636477,
      "loss": 0.3877,
      "step": 1460
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019604893633686662,
      "loss": 0.3822,
      "step": 1470
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019599536043151384,
      "loss": 0.3726,
      "step": 1480
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019594143114753026,
      "loss": 0.3552,
      "step": 1490
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019588714868344073,
      "loss": 0.3725,
      "step": 1500
    },
    {
      "epoch": 0.55,
      "eval_loss": 0.4203263521194458,
      "eval_runtime": 396.7561,
      "eval_samples_per_second": 12.055,
      "eval_steps_per_second": 3.014,
      "step": 1500
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019583251323907006,
      "loss": 0.3993,
      "step": 1510
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001957775250155426,
      "loss": 0.3557,
      "step": 1520
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001957221842152813,
      "loss": 0.4027,
      "step": 1530
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019566649104200696,
      "loss": 0.3547,
      "step": 1540
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019561044570073763,
      "loss": 0.3978,
      "step": 1550
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019555404839778767,
      "loss": 0.3919,
      "step": 1560
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019549729934076717,
      "loss": 0.3713,
      "step": 1570
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019544019873858102,
      "loss": 0.408,
      "step": 1580
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019538274680142834,
      "loss": 0.3792,
      "step": 1590
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019532494374080144,
      "loss": 0.3836,
      "step": 1600
    },
    {
      "epoch": 0.59,
      "eval_loss": 0.42036038637161255,
      "eval_runtime": 402.4927,
      "eval_samples_per_second": 11.883,
      "eval_steps_per_second": 2.971,
      "step": 1600
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019526678976948525,
      "loss": 0.3411,
      "step": 1610
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001952082851015565,
      "loss": 0.3863,
      "step": 1620
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019514942995238287,
      "loss": 0.375,
      "step": 1630
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019509022453862226,
      "loss": 0.3747,
      "step": 1640
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019503066907822198,
      "loss": 0.36,
      "step": 1650
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00019497076379041786,
      "loss": 0.3919,
      "step": 1660
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00019491050889573357,
      "loss": 0.3699,
      "step": 1670
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00019484990461597978,
      "loss": 0.3829,
      "step": 1680
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00019478895117425323,
      "loss": 0.3589,
      "step": 1690
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00019472764879493616,
      "loss": 0.3784,
      "step": 1700
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.4183010458946228,
      "eval_runtime": 404.1371,
      "eval_samples_per_second": 11.835,
      "eval_steps_per_second": 2.959,
      "step": 1700
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00019466599770369509,
      "loss": 0.3912,
      "step": 1710
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00019460399812748041,
      "loss": 0.374,
      "step": 1720
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001945416502945253,
      "loss": 0.384,
      "step": 1730
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00019447895443434486,
      "loss": 0.3519,
      "step": 1740
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00019441591077773554,
      "loss": 0.3598,
      "step": 1750
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001943525195567739,
      "loss": 0.3145,
      "step": 1760
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00019428878100481606,
      "loss": 0.3842,
      "step": 1770
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001942246953564967,
      "loss": 0.3674,
      "step": 1780
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019416026284772825,
      "loss": 0.3603,
      "step": 1790
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019409548371570007,
      "loss": 0.369,
      "step": 1800
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.4111176133155823,
      "eval_runtime": 402.9829,
      "eval_samples_per_second": 11.869,
      "eval_steps_per_second": 2.968,
      "step": 1800
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019403035819887734,
      "loss": 0.3713,
      "step": 1810
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00019396488653700055,
      "loss": 0.3539,
      "step": 1820
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00019389906897108428,
      "loss": 0.3932,
      "step": 1830
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001938329057434165,
      "loss": 0.3371,
      "step": 1840
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00019376639709755766,
      "loss": 0.3832,
      "step": 1850
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00019369954327833972,
      "loss": 0.3658,
      "step": 1860
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00019363234453186534,
      "loss": 0.3452,
      "step": 1870
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00019356480110550687,
      "loss": 0.3721,
      "step": 1880
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00019349691324790555,
      "loss": 0.3637,
      "step": 1890
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00019342868120897054,
      "loss": 0.3409,
      "step": 1900
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.41198766231536865,
      "eval_runtime": 402.9688,
      "eval_samples_per_second": 11.869,
      "eval_steps_per_second": 2.968,
      "step": 1900
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00019336010523987796,
      "loss": 0.3424,
      "step": 1910
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00019329118559307,
      "loss": 0.3532,
      "step": 1920
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001932219225222541,
      "loss": 0.3591,
      "step": 1930
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00019315231628240178,
      "loss": 0.3794,
      "step": 1940
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00019308236712974795,
      "loss": 0.3825,
      "step": 1950
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001930120753217898,
      "loss": 0.3482,
      "step": 1960
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019294144111728584,
      "loss": 0.3643,
      "step": 1970
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019287046477625515,
      "loss": 0.3522,
      "step": 1980
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019279914655997619,
      "loss": 0.3723,
      "step": 1990
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019272748673098596,
      "loss": 0.388,
      "step": 2000
    },
    {
      "epoch": 0.73,
      "eval_loss": 0.41505494713783264,
      "eval_runtime": 404.7184,
      "eval_samples_per_second": 11.818,
      "eval_steps_per_second": 2.955,
      "step": 2000
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.000192655485553079,
      "loss": 0.3739,
      "step": 2010
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00019258314329130641,
      "loss": 0.3625,
      "step": 2020
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00019251046021197496,
      "loss": 0.3174,
      "step": 2030
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019243743658264593,
      "loss": 0.347,
      "step": 2040
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019236407267213433,
      "loss": 0.3837,
      "step": 2050
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019229036875050777,
      "loss": 0.3854,
      "step": 2060
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001922163250890855,
      "loss": 0.3449,
      "step": 2070
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00019214194196043741,
      "loss": 0.3507,
      "step": 2080
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00019206721963838317,
      "loss": 0.3786,
      "step": 2090
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00019199215839799092,
      "loss": 0.3608,
      "step": 2100
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.4092780351638794,
      "eval_runtime": 403.6583,
      "eval_samples_per_second": 11.849,
      "eval_steps_per_second": 2.963,
      "step": 2100
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001919167585155765,
      "loss": 0.3233,
      "step": 2110
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00019184102026870235,
      "loss": 0.3679,
      "step": 2120
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001917649439361765,
      "loss": 0.3426,
      "step": 2130
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00019168852979805162,
      "loss": 0.3272,
      "step": 2140
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00019161177813562379,
      "loss": 0.3572,
      "step": 2150
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001915346892314316,
      "loss": 0.3429,
      "step": 2160
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001914572633692552,
      "loss": 0.3325,
      "step": 2170
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00019137950083411505,
      "loss": 0.3444,
      "step": 2180
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.000191301401912271,
      "loss": 0.351,
      "step": 2190
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00019122296689122123,
      "loss": 0.3171,
      "step": 2200
    },
    {
      "epoch": 0.81,
      "eval_loss": 0.4078885316848755,
      "eval_runtime": 403.041,
      "eval_samples_per_second": 11.867,
      "eval_steps_per_second": 2.967,
      "step": 2200
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001911441960597012,
      "loss": 0.3626,
      "step": 2210
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001910650897076824,
      "loss": 0.3529,
      "step": 2220
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00019098564812637165,
      "loss": 0.3727,
      "step": 2230
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001909058716082097,
      "loss": 0.3111,
      "step": 2240
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001908257604468703,
      "loss": 0.3846,
      "step": 2250
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00019074531493725906,
      "loss": 0.3612,
      "step": 2260
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001906645353755124,
      "loss": 0.3573,
      "step": 2270
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00019058342205899656,
      "loss": 0.3902,
      "step": 2280
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001905019752863062,
      "loss": 0.345,
      "step": 2290
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001904201953572637,
      "loss": 0.3581,
      "step": 2300
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.4118511974811554,
      "eval_runtime": 405.9796,
      "eval_samples_per_second": 11.781,
      "eval_steps_per_second": 2.946,
      "step": 2300
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00019033808257291768,
      "loss": 0.3303,
      "step": 2310
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00019025563723554223,
      "loss": 0.3432,
      "step": 2320
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00019017285964863554,
      "loss": 0.3436,
      "step": 2330
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001900897501169189,
      "loss": 0.338,
      "step": 2340
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001900063089463356,
      "loss": 0.3611,
      "step": 2350
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00018992253644404967,
      "loss": 0.3311,
      "step": 2360
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018983843291844492,
      "loss": 0.3337,
      "step": 2370
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018975399867912364,
      "loss": 0.358,
      "step": 2380
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001896692340369057,
      "loss": 0.3802,
      "step": 2390
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00018958413930382704,
      "loss": 0.3389,
      "step": 2400
    },
    {
      "epoch": 0.88,
      "eval_loss": 0.41193756461143494,
      "eval_runtime": 402.9836,
      "eval_samples_per_second": 11.869,
      "eval_steps_per_second": 2.968,
      "step": 2400
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001894987147931389,
      "loss": 0.3398,
      "step": 2410
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00018941296081930646,
      "loss": 0.3459,
      "step": 2420
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00018932687769800767,
      "loss": 0.3546,
      "step": 2430
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00018924046574613222,
      "loss": 0.3158,
      "step": 2440
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001891537252817802,
      "loss": 0.3484,
      "step": 2450
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00018906665662426104,
      "loss": 0.3437,
      "step": 2460
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001889792600940924,
      "loss": 0.3637,
      "step": 2470
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00018889153601299888,
      "loss": 0.3658,
      "step": 2480
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00018880348470391077,
      "loss": 0.3452,
      "step": 2490
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.000188715106490963,
      "loss": 0.3302,
      "step": 2500
    },
    {
      "epoch": 0.92,
      "eval_loss": 0.4021553695201874,
      "eval_runtime": 399.2274,
      "eval_samples_per_second": 11.981,
      "eval_steps_per_second": 2.996,
      "step": 2500
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.000188626401699494,
      "loss": 0.3511,
      "step": 2510
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00018853737065604426,
      "loss": 0.3232,
      "step": 2520
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00018844801368835532,
      "loss": 0.366,
      "step": 2530
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00018835833112536857,
      "loss": 0.3151,
      "step": 2540
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001882683232972239,
      "loss": 0.3534,
      "step": 2550
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00018817799053525862,
      "loss": 0.3779,
      "step": 2560
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001880873331720062,
      "loss": 0.3338,
      "step": 2570
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00018799635154119495,
      "loss": 0.3243,
      "step": 2580
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00018790504597774698,
      "loss": 0.3729,
      "step": 2590
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001878134168177768,
      "loss": 0.3553,
      "step": 2600
    },
    {
      "epoch": 0.95,
      "eval_loss": 0.4054717421531677,
      "eval_runtime": 400.3737,
      "eval_samples_per_second": 11.946,
      "eval_steps_per_second": 2.987,
      "step": 2600
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00018772146439859015,
      "loss": 0.3268,
      "step": 2610
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00018762918905868277,
      "loss": 0.3304,
      "step": 2620
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00018753659113773913,
      "loss": 0.3606,
      "step": 2630
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001874436709766312,
      "loss": 0.3421,
      "step": 2640
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00018735042891741718,
      "loss": 0.3629,
      "step": 2650
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001872568653033402,
      "loss": 0.3636,
      "step": 2660
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00018716298047882714,
      "loss": 0.3485,
      "step": 2670
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00018706877478948735,
      "loss": 0.3169,
      "step": 2680
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00018697424858211126,
      "loss": 0.306,
      "step": 2690
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001868794022046693,
      "loss": 0.3586,
      "step": 2700
    },
    {
      "epoch": 0.99,
      "eval_loss": 0.4048784375190735,
      "eval_runtime": 400.9367,
      "eval_samples_per_second": 11.93,
      "eval_steps_per_second": 2.983,
      "step": 2700
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00018678423600631042,
      "loss": 0.3311,
      "step": 2710
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00018668875033736094,
      "loss": 0.3066,
      "step": 2720
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00018659294554932324,
      "loss": 0.338,
      "step": 2730
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00018649682199487437,
      "loss": 0.3205,
      "step": 2740
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001864003800278649,
      "loss": 0.3176,
      "step": 2750
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00018630362000331753,
      "loss": 0.2986,
      "step": 2760
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00018620654227742572,
      "loss": 0.3175,
      "step": 2770
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001861091472075526,
      "loss": 0.3174,
      "step": 2780
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0001860114351522293,
      "loss": 0.2752,
      "step": 2790
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00018591340647115402,
      "loss": 0.3014,
      "step": 2800
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.41257622838020325,
      "eval_runtime": 223.7363,
      "eval_samples_per_second": 21.378,
      "eval_steps_per_second": 5.346,
      "step": 2800
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001858150615251905,
      "loss": 0.3013,
      "step": 2810
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00018571640067636662,
      "loss": 0.3364,
      "step": 2820
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00018561742428787324,
      "loss": 0.3159,
      "step": 2830
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00018552807605292504,
      "loss": 0.3037,
      "step": 2840
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00018542850114380946,
      "loss": 0.3087,
      "step": 2850
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00018532861175484162,
      "loss": 0.2858,
      "step": 2860
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00018522840825373492,
      "loss": 0.3208,
      "step": 2870
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00018512789100935906,
      "loss": 0.3117,
      "step": 2880
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00018502706039173856,
      "loss": 0.2921,
      "step": 2890
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001849259167720517,
      "loss": 0.3354,
      "step": 2900
    },
    {
      "epoch": 1.06,
      "eval_loss": 0.4081941545009613,
      "eval_runtime": 223.8258,
      "eval_samples_per_second": 21.369,
      "eval_steps_per_second": 5.343,
      "step": 2900
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001848244605226289,
      "loss": 0.2866,
      "step": 2910
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001847226920169514,
      "loss": 0.308,
      "step": 2920
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00018462061162965,
      "loss": 0.3142,
      "step": 2930
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001845182197365036,
      "loss": 0.3125,
      "step": 2940
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00018441551671443768,
      "loss": 0.307,
      "step": 2950
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00018431250294152323,
      "loss": 0.2804,
      "step": 2960
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00018420917879697507,
      "loss": 0.304,
      "step": 2970
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001841055446611506,
      "loss": 0.3152,
      "step": 2980
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00018400160091554835,
      "loss": 0.3095,
      "step": 2990
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001838973479428066,
      "loss": 0.2954,
      "step": 3000
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.4158288836479187,
      "eval_runtime": 223.8063,
      "eval_samples_per_second": 21.371,
      "eval_steps_per_second": 5.344,
      "step": 3000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00018379278612670193,
      "loss": 0.3167,
      "step": 3010
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00018368791585214784,
      "loss": 0.2872,
      "step": 3020
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00018358273750519337,
      "loss": 0.2999,
      "step": 3030
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00018347725147302158,
      "loss": 0.3229,
      "step": 3040
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00018337145814394825,
      "loss": 0.3332,
      "step": 3050
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001832653579074203,
      "loss": 0.2818,
      "step": 3060
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00018315895115401457,
      "loss": 0.3198,
      "step": 3070
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00018305223827543604,
      "loss": 0.3504,
      "step": 3080
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001829452196645168,
      "loss": 0.3392,
      "step": 3090
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00018283789571521436,
      "loss": 0.3023,
      "step": 3100
    },
    {
      "epoch": 1.14,
      "eval_loss": 0.40502193570137024,
      "eval_runtime": 223.8877,
      "eval_samples_per_second": 21.363,
      "eval_steps_per_second": 5.342,
      "step": 3100
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00018273026682261013,
      "loss": 0.3164,
      "step": 3110
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001826223333829082,
      "loss": 0.2985,
      "step": 3120
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00018251409579343375,
      "loss": 0.2992,
      "step": 3130
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00018240555445263153,
      "loss": 0.3174,
      "step": 3140
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00018229670976006453,
      "loss": 0.3025,
      "step": 3150
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00018218756211641236,
      "loss": 0.3297,
      "step": 3160
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00018207811192346996,
      "loss": 0.3139,
      "step": 3170
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00018196835958414598,
      "loss": 0.2956,
      "step": 3180
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00018185830550246124,
      "loss": 0.3189,
      "step": 3190
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00018174795008354743,
      "loss": 0.2896,
      "step": 3200
    },
    {
      "epoch": 1.17,
      "eval_loss": 0.40529006719589233,
      "eval_runtime": 223.8738,
      "eval_samples_per_second": 21.365,
      "eval_steps_per_second": 5.342,
      "step": 3200
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00018163729373364554,
      "loss": 0.3085,
      "step": 3210
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001815263368601043,
      "loss": 0.316,
      "step": 3220
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00018141507987137873,
      "loss": 0.326,
      "step": 3230
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00018130352317702865,
      "loss": 0.293,
      "step": 3240
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00018119166718771716,
      "loss": 0.2887,
      "step": 3250
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00018107951231520911,
      "loss": 0.3266,
      "step": 3260
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00018096705897236966,
      "loss": 0.3327,
      "step": 3270
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00018085430757316256,
      "loss": 0.2991,
      "step": 3280
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00018074125853264898,
      "loss": 0.3076,
      "step": 3290
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00018062791226698558,
      "loss": 0.3339,
      "step": 3300
    },
    {
      "epoch": 1.21,
      "eval_loss": 0.4053775668144226,
      "eval_runtime": 223.9954,
      "eval_samples_per_second": 21.353,
      "eval_steps_per_second": 5.339,
      "step": 3300
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00018051426919342317,
      "loss": 0.2935,
      "step": 3310
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00018040032973030536,
      "loss": 0.3497,
      "step": 3320
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00018028609429706664,
      "loss": 0.2897,
      "step": 3330
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00018017156331423114,
      "loss": 0.312,
      "step": 3340
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00018005673720341086,
      "loss": 0.303,
      "step": 3350
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00017994161638730432,
      "loss": 0.3253,
      "step": 3360
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00017982620128969488,
      "loss": 0.3322,
      "step": 3370
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001797104923354492,
      "loss": 0.2988,
      "step": 3380
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00017959448995051575,
      "loss": 0.3094,
      "step": 3390
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00017947819456192306,
      "loss": 0.3118,
      "step": 3400
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.3964312672615051,
      "eval_runtime": 223.8688,
      "eval_samples_per_second": 21.365,
      "eval_steps_per_second": 5.342,
      "step": 3400
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00017936160659777833,
      "loss": 0.2958,
      "step": 3410
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00017924472648726583,
      "loss": 0.3119,
      "step": 3420
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00017912755466064525,
      "loss": 0.3096,
      "step": 3430
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00017901009154925007,
      "loss": 0.2778,
      "step": 3440
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00017889233758548625,
      "loss": 0.2953,
      "step": 3450
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00017877429320283016,
      "loss": 0.3063,
      "step": 3460
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001786559588358275,
      "loss": 0.2945,
      "step": 3470
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00017853733492009135,
      "loss": 0.3027,
      "step": 3480
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001784184218923007,
      "loss": 0.2654,
      "step": 3490
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001782992201901988,
      "loss": 0.3289,
      "step": 3500
    },
    {
      "epoch": 1.28,
      "eval_loss": 0.39914268255233765,
      "eval_runtime": 223.8716,
      "eval_samples_per_second": 21.365,
      "eval_steps_per_second": 5.342,
      "step": 3500
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001781797302525916,
      "loss": 0.2894,
      "step": 3510
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00017805995251934614,
      "loss": 0.2776,
      "step": 3520
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00017793988743138877,
      "loss": 0.288,
      "step": 3530
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00017781953543070372,
      "loss": 0.3229,
      "step": 3540
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00017769889696033154,
      "loss": 0.285,
      "step": 3550
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001775779724643671,
      "loss": 0.321,
      "step": 3560
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001774567623879583,
      "loss": 0.3222,
      "step": 3570
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00017733526717730435,
      "loss": 0.3087,
      "step": 3580
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00017721348727965408,
      "loss": 0.2904,
      "step": 3590
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00017709142314330424,
      "loss": 0.2984,
      "step": 3600
    },
    {
      "epoch": 1.32,
      "eval_loss": 0.40588298439979553,
      "eval_runtime": 223.9372,
      "eval_samples_per_second": 21.359,
      "eval_steps_per_second": 5.341,
      "step": 3600
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00017696907521759804,
      "loss": 0.3216,
      "step": 3610
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00017684644395292326,
      "loss": 0.3019,
      "step": 3620
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00017672352980071078,
      "loss": 0.3272,
      "step": 3630
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00017660033321343285,
      "loss": 0.2892,
      "step": 3640
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001764768546446014,
      "loss": 0.2921,
      "step": 3650
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00017635309454876636,
      "loss": 0.3105,
      "step": 3660
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00017622905338151408,
      "loss": 0.3205,
      "step": 3670
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00017610473159946556,
      "loss": 0.2838,
      "step": 3680
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00017598012966027482,
      "loss": 0.2762,
      "step": 3690
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001758552480226271,
      "loss": 0.3277,
      "step": 3700
    },
    {
      "epoch": 1.36,
      "eval_loss": 0.3979549705982208,
      "eval_runtime": 223.8768,
      "eval_samples_per_second": 21.364,
      "eval_steps_per_second": 5.342,
      "step": 3700
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00017573008714623746,
      "loss": 0.2709,
      "step": 3710
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00017560464749184876,
      "loss": 0.2918,
      "step": 3720
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00017547892952123005,
      "loss": 0.3098,
      "step": 3730
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00017535293369717506,
      "loss": 0.3233,
      "step": 3740
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00017522666048350023,
      "loss": 0.31,
      "step": 3750
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00017510011034504324,
      "loss": 0.3337,
      "step": 3760
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00017497328374766112,
      "loss": 0.2845,
      "step": 3770
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00017484618115822857,
      "loss": 0.3073,
      "step": 3780
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00017471880304463638,
      "loss": 0.2977,
      "step": 3790
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001745911498757895,
      "loss": 0.3011,
      "step": 3800
    },
    {
      "epoch": 1.39,
      "eval_loss": 0.4045478105545044,
      "eval_runtime": 223.7983,
      "eval_samples_per_second": 21.372,
      "eval_steps_per_second": 5.344,
      "step": 3800
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00017446322212160545,
      "loss": 0.3029,
      "step": 3810
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001743350202530126,
      "loss": 0.3017,
      "step": 3820
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00017420654474194832,
      "loss": 0.3179,
      "step": 3830
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00017407779606135732,
      "loss": 0.2728,
      "step": 3840
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00017394877468518996,
      "loss": 0.286,
      "step": 3850
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00017381948108840042,
      "loss": 0.3009,
      "step": 3860
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00017368991574694495,
      "loss": 0.2946,
      "step": 3870
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001735600791377802,
      "loss": 0.3229,
      "step": 3880
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00017342997173886134,
      "loss": 0.3372,
      "step": 3890
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00017329959402914046,
      "loss": 0.3194,
      "step": 3900
    },
    {
      "epoch": 1.43,
      "eval_loss": 0.40152063965797424,
      "eval_runtime": 223.8706,
      "eval_samples_per_second": 21.365,
      "eval_steps_per_second": 5.342,
      "step": 3900
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001731689464885647,
      "loss": 0.2674,
      "step": 3910
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00017303802959807443,
      "loss": 0.2558,
      "step": 3920
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001729068438396016,
      "loss": 0.3186,
      "step": 3930
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00017277538969606793,
      "loss": 0.3298,
      "step": 3940
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00017264366765138317,
      "loss": 0.2991,
      "step": 3950
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00017251167819044315,
      "loss": 0.2652,
      "step": 3960
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001723794217991282,
      "loss": 0.3177,
      "step": 3970
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00017224689896430117,
      "loss": 0.3045,
      "step": 3980
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00017211411017380594,
      "loss": 0.3146,
      "step": 3990
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00017198105591646528,
      "loss": 0.2921,
      "step": 4000
    },
    {
      "epoch": 1.47,
      "eval_loss": 0.400907427072525,
      "eval_runtime": 224.0601,
      "eval_samples_per_second": 21.347,
      "eval_steps_per_second": 5.338,
      "step": 4000
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00017184773668207917,
      "loss": 0.3441,
      "step": 4010
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00017171415296142315,
      "loss": 0.2876,
      "step": 4020
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00017158030524624626,
      "loss": 0.2755,
      "step": 4030
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001714461940292695,
      "loss": 0.3169,
      "step": 4040
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00017131181980418374,
      "loss": 0.2844,
      "step": 4050
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00017117718306564812,
      "loss": 0.2821,
      "step": 4060
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00017104228430928805,
      "loss": 0.3002,
      "step": 4070
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00017090712403169364,
      "loss": 0.3232,
      "step": 4080
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00017077170273041757,
      "loss": 0.3135,
      "step": 4090
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00017063602090397346,
      "loss": 0.2917,
      "step": 4100
    },
    {
      "epoch": 1.5,
      "eval_loss": 0.4019235670566559,
      "eval_runtime": 224.3869,
      "eval_samples_per_second": 21.316,
      "eval_steps_per_second": 5.33,
      "step": 4100
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00017050007905183398,
      "loss": 0.3089,
      "step": 4110
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.000170363877674429,
      "loss": 0.3063,
      "step": 4120
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00017022741727314373,
      "loss": 0.3056,
      "step": 4130
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00017009069835031694,
      "loss": 0.3168,
      "step": 4140
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00016995372140923907,
      "loss": 0.3062,
      "step": 4150
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00016981648695415033,
      "loss": 0.2744,
      "step": 4160
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00016967899549023895,
      "loss": 0.2877,
      "step": 4170
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00016954124752363922,
      "loss": 0.3173,
      "step": 4180
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00016940324356142972,
      "loss": 0.2954,
      "step": 4190
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00016926498411163135,
      "loss": 0.2792,
      "step": 4200
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.4046282172203064,
      "eval_runtime": 224.0328,
      "eval_samples_per_second": 21.35,
      "eval_steps_per_second": 5.339,
      "step": 4200
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00016912646968320552,
      "loss": 0.3168,
      "step": 4210
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00016898770078605226,
      "loss": 0.2854,
      "step": 4220
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00016884867793100843,
      "loss": 0.2979,
      "step": 4230
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00016870940162984566,
      "loss": 0.2656,
      "step": 4240
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00016856987239526863,
      "loss": 0.3169,
      "step": 4250
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00016843009074091306,
      "loss": 0.2872,
      "step": 4260
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00016829005718134397,
      "loss": 0.2954,
      "step": 4270
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00016814977223205362,
      "loss": 0.2885,
      "step": 4280
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00016800923640945974,
      "loss": 0.2927,
      "step": 4290
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0001678684502309035,
      "loss": 0.2886,
      "step": 4300
    },
    {
      "epoch": 1.58,
      "eval_loss": 0.4055146872997284,
      "eval_runtime": 224.1142,
      "eval_samples_per_second": 21.342,
      "eval_steps_per_second": 5.337,
      "step": 4300
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00016772741421464772,
      "loss": 0.3042,
      "step": 4310
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00016758612887987498,
      "loss": 0.2858,
      "step": 4320
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00016744459474668557,
      "loss": 0.3072,
      "step": 4330
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00016730281233609572,
      "loss": 0.3126,
      "step": 4340
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00016716078217003557,
      "loss": 0.2748,
      "step": 4350
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00016701850477134734,
      "loss": 0.2873,
      "step": 4360
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00016687598066378336,
      "loss": 0.3051,
      "step": 4370
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00016673321037200407,
      "loss": 0.3077,
      "step": 4380
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001665901944215763,
      "loss": 0.2708,
      "step": 4390
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00016644693333897108,
      "loss": 0.2947,
      "step": 4400
    },
    {
      "epoch": 1.61,
      "eval_loss": 0.40514013171195984,
      "eval_runtime": 224.1094,
      "eval_samples_per_second": 21.342,
      "eval_steps_per_second": 5.337,
      "step": 4400
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001663034276515619,
      "loss": 0.306,
      "step": 4410
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00016615967788762261,
      "loss": 0.3076,
      "step": 4420
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00016601568457632566,
      "loss": 0.2895,
      "step": 4430
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00016587144824773992,
      "loss": 0.2885,
      "step": 4440
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00016572696943282903,
      "loss": 0.3046,
      "step": 4450
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00016558224866344907,
      "loss": 0.3037,
      "step": 4460
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.000165437286472347,
      "loss": 0.28,
      "step": 4470
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00016529208339315833,
      "loss": 0.2903,
      "step": 4480
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00016514663996040544,
      "loss": 0.2849,
      "step": 4490
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00016500095670949548,
      "loss": 0.2975,
      "step": 4500
    },
    {
      "epoch": 1.65,
      "eval_loss": 0.4067119359970093,
      "eval_runtime": 224.4914,
      "eval_samples_per_second": 21.306,
      "eval_steps_per_second": 5.328,
      "step": 4500
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00016485503417671836,
      "loss": 0.3115,
      "step": 4510
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00016470887289924492,
      "loss": 0.3409,
      "step": 4520
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00016456247341512485,
      "loss": 0.2604,
      "step": 4530
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00016441583626328467,
      "loss": 0.3226,
      "step": 4540
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00016426896198352587,
      "loss": 0.2816,
      "step": 4550
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00016412185111652278,
      "loss": 0.319,
      "step": 4560
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00016397450420382076,
      "loss": 0.2831,
      "step": 4570
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00016382692178783402,
      "loss": 0.2974,
      "step": 4580
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00016367910441184374,
      "loss": 0.2772,
      "step": 4590
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00016353105261999605,
      "loss": 0.3091,
      "step": 4600
    },
    {
      "epoch": 1.69,
      "eval_loss": 0.3946963846683502,
      "eval_runtime": 224.3647,
      "eval_samples_per_second": 21.318,
      "eval_steps_per_second": 5.331,
      "step": 4600
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00016338276695729994,
      "loss": 0.286,
      "step": 4610
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00016323424796962544,
      "loss": 0.2721,
      "step": 4620
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001630854962037014,
      "loss": 0.2983,
      "step": 4630
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00016293651220711364,
      "loss": 0.2875,
      "step": 4640
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001627872965283028,
      "loss": 0.2859,
      "step": 4650
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00016263784971656247,
      "loss": 0.2603,
      "step": 4660
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00016248817232203698,
      "loss": 0.3231,
      "step": 4670
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00016233826489571963,
      "loss": 0.2927,
      "step": 4680
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00016218812798945038,
      "loss": 0.297,
      "step": 4690
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00016203776215591403,
      "loss": 0.2908,
      "step": 4700
    },
    {
      "epoch": 1.72,
      "eval_loss": 0.403292179107666,
      "eval_runtime": 224.3888,
      "eval_samples_per_second": 21.316,
      "eval_steps_per_second": 5.33,
      "step": 4700
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001618871679486381,
      "loss": 0.3094,
      "step": 4710
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00016173634592199076,
      "loss": 0.29,
      "step": 4720
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00016158529663117888,
      "loss": 0.3115,
      "step": 4730
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00016143402063224584,
      "loss": 0.3139,
      "step": 4740
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00016128251848206976,
      "loss": 0.302,
      "step": 4750
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00016113079073836107,
      "loss": 0.2777,
      "step": 4760
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001609788379596608,
      "loss": 0.2906,
      "step": 4770
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00016082666070533832,
      "loss": 0.3028,
      "step": 4780
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001606742595355893,
      "loss": 0.3159,
      "step": 4790
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.0001605216350114338,
      "loss": 0.2864,
      "step": 4800
    },
    {
      "epoch": 1.76,
      "eval_loss": 0.40955105423927307,
      "eval_runtime": 224.3793,
      "eval_samples_per_second": 21.317,
      "eval_steps_per_second": 5.33,
      "step": 4800
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00016036878769471401,
      "loss": 0.3017,
      "step": 4810
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00016021571814809227,
      "loss": 0.2986,
      "step": 4820
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.000160062426935049,
      "loss": 0.3077,
      "step": 4830
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00015990891461988065,
      "loss": 0.2758,
      "step": 4840
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00015975518176769755,
      "loss": 0.2592,
      "step": 4850
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001596012289444219,
      "loss": 0.3021,
      "step": 4860
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00015944705671678565,
      "loss": 0.3047,
      "step": 4870
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00015929266565232846,
      "loss": 0.2785,
      "step": 4880
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00015913805631939546,
      "loss": 0.2839,
      "step": 4890
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00015898322928713544,
      "loss": 0.2788,
      "step": 4900
    },
    {
      "epoch": 1.8,
      "eval_loss": 0.40693148970603943,
      "eval_runtime": 224.6917,
      "eval_samples_per_second": 21.287,
      "eval_steps_per_second": 5.323,
      "step": 4900
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00015882818512549836,
      "loss": 0.2875,
      "step": 4910
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00015867292440523378,
      "loss": 0.2907,
      "step": 4920
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00015851744769788818,
      "loss": 0.2776,
      "step": 4930
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00015836175557580324,
      "loss": 0.2608,
      "step": 4940
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00015820584861211368,
      "loss": 0.2945,
      "step": 4950
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.000158049727380745,
      "loss": 0.2771,
      "step": 4960
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001578933924564115,
      "loss": 0.2802,
      "step": 4970
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001577368444146142,
      "loss": 0.2849,
      "step": 4980
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00015758008383163854,
      "loss": 0.3013,
      "step": 4990
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001574231112845524,
      "loss": 0.2942,
      "step": 5000
    },
    {
      "epoch": 1.83,
      "eval_loss": 0.399748831987381,
      "eval_runtime": 224.6973,
      "eval_samples_per_second": 21.286,
      "eval_steps_per_second": 5.323,
      "step": 5000
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00015726592735120393,
      "loss": 0.2883,
      "step": 5010
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001571085326102195,
      "loss": 0.2639,
      "step": 5020
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001569509276410015,
      "loss": 0.2624,
      "step": 5030
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00015679311302372614,
      "loss": 0.2823,
      "step": 5040
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001566350893393414,
      "loss": 0.3026,
      "step": 5050
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00015647685716956494,
      "loss": 0.2803,
      "step": 5060
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00015631841709688184,
      "loss": 0.2991,
      "step": 5070
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00015615976970454257,
      "loss": 0.2734,
      "step": 5080
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00015600091557656072,
      "loss": 0.2674,
      "step": 5090
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.000155841855297711,
      "loss": 0.2736,
      "step": 5100
    },
    {
      "epoch": 1.87,
      "eval_loss": 0.40985968708992004,
      "eval_runtime": 224.6419,
      "eval_samples_per_second": 21.292,
      "eval_steps_per_second": 5.324,
      "step": 5100
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0001556825894535269,
      "loss": 0.2828,
      "step": 5110
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00015552311863029875,
      "loss": 0.2872,
      "step": 5120
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00015536344341507129,
      "loss": 0.2669,
      "step": 5130
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001552035643956419,
      "loss": 0.2708,
      "step": 5140
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00015504348216055798,
      "loss": 0.3048,
      "step": 5150
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00015488319729911512,
      "loss": 0.2964,
      "step": 5160
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00015472271040135483,
      "loss": 0.266,
      "step": 5170
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00015456202205806234,
      "loss": 0.2885,
      "step": 5180
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001544011328607644,
      "loss": 0.3317,
      "step": 5190
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00015424004340172719,
      "loss": 0.2905,
      "step": 5200
    },
    {
      "epoch": 1.91,
      "eval_loss": 0.39825478196144104,
      "eval_runtime": 224.8976,
      "eval_samples_per_second": 21.267,
      "eval_steps_per_second": 5.318,
      "step": 5200
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0001540787542739541,
      "loss": 0.2663,
      "step": 5210
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00015391726607118345,
      "loss": 0.2838,
      "step": 5220
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00015375557938788657,
      "loss": 0.2962,
      "step": 5230
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001535936948192653,
      "loss": 0.2884,
      "step": 5240
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00015343161296124994,
      "loss": 0.3026,
      "step": 5250
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00015326933441049714,
      "loss": 0.3234,
      "step": 5260
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00015310685976438753,
      "loss": 0.2547,
      "step": 5270
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00015294418962102363,
      "loss": 0.2896,
      "step": 5280
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00015278132457922764,
      "loss": 0.3173,
      "step": 5290
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00015261826523853926,
      "loss": 0.2848,
      "step": 5300
    },
    {
      "epoch": 1.94,
      "eval_loss": 0.4000326693058014,
      "eval_runtime": 224.8241,
      "eval_samples_per_second": 21.274,
      "eval_steps_per_second": 5.32,
      "step": 5300
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00015245501219921336,
      "loss": 0.2819,
      "step": 5310
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00015229156606221792,
      "loss": 0.2578,
      "step": 5320
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001521279274292317,
      "loss": 0.2893,
      "step": 5330
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00015196409690264212,
      "loss": 0.2897,
      "step": 5340
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.000151800075085543,
      "loss": 0.2718,
      "step": 5350
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001516358625817323,
      "loss": 0.2908,
      "step": 5360
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00015147145999570998,
      "loss": 0.2694,
      "step": 5370
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0001513068679326757,
      "loss": 0.3028,
      "step": 5380
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00015114208699852663,
      "loss": 0.2707,
      "step": 5390
    },
    {
      "epoch": 1.98,
      "step": 5391,
      "total_flos": 1.7572375333880463e+18,
      "train_loss": 4.6051067095080594e-05,
      "train_runtime": 5.408,
      "train_samples_per_second": 15680.552,
      "train_steps_per_second": 980.035
    }
  ],
  "logging_steps": 10,
  "max_steps": 5300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 10,
  "total_flos": 1.7572375333880463e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}