{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8920877793436682,
  "eval_steps": 100,
  "global_step": 4431,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.0001999999444378697,
      "loss": 0.7972,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0001999997777515405,
      "loss": 0.221,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001999994999411976,
      "loss": 0.1313,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019999911100714978,
      "loss": 0.1169,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019999861094982923,
      "loss": 0.1353,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019999799976979157,
      "loss": 0.1355,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019999727746771603,
      "loss": 0.0991,
      "step": 70
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019999644404440528,
      "loss": 0.0995,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019999549950078536,
      "loss": 0.1175,
      "step": 90
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000199994443837906,
      "loss": 0.0941,
      "step": 100
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.09952569007873535,
      "eval_runtime": 659.9109,
      "eval_samples_per_second": 11.15,
      "eval_steps_per_second": 2.788,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019999327705694028,
      "loss": 0.09,
      "step": 110
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019999199915918467,
      "loss": 0.1044,
      "step": 120
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019999061014605934,
      "loss": 0.0978,
      "step": 130
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019998911001910778,
      "loss": 0.0963,
      "step": 140
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000199987498779997,
      "loss": 0.0955,
      "step": 150
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019998577643051744,
      "loss": 0.0994,
      "step": 160
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019998394297258311,
      "loss": 0.0921,
      "step": 170
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001999819984082314,
      "loss": 0.087,
      "step": 180
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019997994273962316,
      "loss": 0.0826,
      "step": 190
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019997777596904282,
      "loss": 0.0903,
      "step": 200
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.090216726064682,
      "eval_runtime": 649.064,
      "eval_samples_per_second": 11.336,
      "eval_steps_per_second": 2.835,
      "step": 200
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019997549809889811,
      "loss": 0.0949,
      "step": 210
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019997310913172028,
      "loss": 0.0902,
      "step": 220
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019997060907016415,
      "loss": 0.0792,
      "step": 230
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001999679979170078,
      "loss": 0.0901,
      "step": 240
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001999652756751529,
      "loss": 0.1035,
      "step": 250
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019996244234762452,
      "loss": 0.0875,
      "step": 260
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019995949793757118,
      "loss": 0.0718,
      "step": 270
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019995644244826481,
      "loss": 0.0799,
      "step": 280
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001999532758831008,
      "loss": 0.0767,
      "step": 290
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000199949998245598,
      "loss": 0.089,
      "step": 300
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.08015166223049164,
      "eval_runtime": 646.37,
      "eval_samples_per_second": 11.384,
      "eval_steps_per_second": 2.847,
      "step": 300
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019994660953939864,
      "loss": 0.0833,
      "step": 310
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019994310976826837,
      "loss": 0.0799,
      "step": 320
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019993949893609635,
      "loss": 0.0747,
      "step": 330
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019993577704689505,
      "loss": 0.0888,
      "step": 340
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001999319441048004,
      "loss": 0.0872,
      "step": 350
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019992800011407172,
      "loss": 0.0762,
      "step": 360
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019992394507909174,
      "loss": 0.0746,
      "step": 370
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019991977900436658,
      "loss": 0.0713,
      "step": 380
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001999155018945258,
      "loss": 0.077,
      "step": 390
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019991111375432225,
      "loss": 0.0619,
      "step": 400
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.07623383402824402,
      "eval_runtime": 654.0144,
      "eval_samples_per_second": 11.251,
      "eval_steps_per_second": 2.813,
      "step": 400
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019990661458863226,
      "loss": 0.0702,
      "step": 410
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001999020044024555,
      "loss": 0.0725,
      "step": 420
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019989728320091496,
      "loss": 0.065,
      "step": 430
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019989245098925708,
      "loss": 0.0632,
      "step": 440
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019988750777285162,
      "loss": 0.0762,
      "step": 450
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019988245355719166,
      "loss": 0.0736,
      "step": 460
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001998772883478937,
      "loss": 0.0662,
      "step": 470
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001998720121506975,
      "loss": 0.0765,
      "step": 480
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019986662497146622,
      "loss": 0.0694,
      "step": 490
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019986112681618634,
      "loss": 0.0722,
      "step": 500
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.075162373483181,
      "eval_runtime": 656.2929,
      "eval_samples_per_second": 11.211,
      "eval_steps_per_second": 2.804,
      "step": 500
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001998555176909676,
      "loss": 0.073,
      "step": 510
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019984979760204313,
      "loss": 0.0681,
      "step": 520
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019984396655576932,
      "loss": 0.057,
      "step": 530
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001998380245586259,
      "loss": 0.0689,
      "step": 540
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019983197161721583,
      "loss": 0.057,
      "step": 550
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019982580773826545,
      "loss": 0.0693,
      "step": 560
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019981953292862428,
      "loss": 0.0702,
      "step": 570
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001998131471952652,
      "loss": 0.0681,
      "step": 580
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019980665054528425,
      "loss": 0.0657,
      "step": 590
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019980004298590081,
      "loss": 0.0724,
      "step": 600
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.0736309364438057,
      "eval_runtime": 657.9967,
      "eval_samples_per_second": 11.182,
      "eval_steps_per_second": 2.796,
      "step": 600
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019979332452445752,
      "loss": 0.0599,
      "step": 610
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019978649516842016,
      "loss": 0.0697,
      "step": 620
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019977955492537787,
      "loss": 0.0715,
      "step": 630
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019977250380304287,
      "loss": 0.0589,
      "step": 640
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001997653418092507,
      "loss": 0.0722,
      "step": 650
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019975806895196008,
      "loss": 0.0611,
      "step": 660
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001997506852392529,
      "loss": 0.056,
      "step": 670
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001997431906793343,
      "loss": 0.0614,
      "step": 680
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019973558528053252,
      "loss": 0.0537,
      "step": 690
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.000199727869051299,
      "loss": 0.0762,
      "step": 700
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.07101555913686752,
      "eval_runtime": 665.8897,
      "eval_samples_per_second": 11.05,
      "eval_steps_per_second": 2.763,
      "step": 700
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019972004200020832,
      "loss": 0.0634,
      "step": 710
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001997121041359583,
      "loss": 0.0697,
      "step": 720
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019970405546736978,
      "loss": 0.0605,
      "step": 730
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019969589600338678,
      "loss": 0.0598,
      "step": 740
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019968762575307649,
      "loss": 0.066,
      "step": 750
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019967924472562914,
      "loss": 0.0656,
      "step": 760
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019967075293035805,
      "loss": 0.0614,
      "step": 770
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001996621503766997,
      "loss": 0.0633,
      "step": 780
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019965343707421362,
      "loss": 0.0661,
      "step": 790
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019964461303258242,
      "loss": 0.0567,
      "step": 800
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.07119747996330261,
      "eval_runtime": 657.3478,
      "eval_samples_per_second": 11.193,
      "eval_steps_per_second": 2.799,
      "step": 800
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019963567826161168,
      "loss": 0.0577,
      "step": 810
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019962663277123016,
      "loss": 0.0633,
      "step": 820
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001996174765714896,
      "loss": 0.0701,
      "step": 830
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001996082096725647,
      "loss": 0.0715,
      "step": 840
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001995988320847533,
      "loss": 0.0618,
      "step": 850
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019958934381847612,
      "loss": 0.0589,
      "step": 860
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019957974488427698,
      "loss": 0.0658,
      "step": 870
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019957003529282255,
      "loss": 0.0549,
      "step": 880
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019956021505490262,
      "loss": 0.0664,
      "step": 890
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001995502841814298,
      "loss": 0.0578,
      "step": 900
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.06790292263031006,
      "eval_runtime": 663.3404,
      "eval_samples_per_second": 11.092,
      "eval_steps_per_second": 2.774,
      "step": 900
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019954024268343975,
      "loss": 0.0626,
      "step": 910
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.000199530090572091,
      "loss": 0.0621,
      "step": 920
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019951982785866492,
      "loss": 0.0527,
      "step": 930
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019950945455456597,
      "loss": 0.0636,
      "step": 940
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019949897067132142,
      "loss": 0.0767,
      "step": 950
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019948837622058135,
      "loss": 0.0785,
      "step": 960
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019947767121411878,
      "loss": 0.0698,
      "step": 970
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001994668556638296,
      "loss": 0.0576,
      "step": 980
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019945592958173247,
      "loss": 0.0616,
      "step": 990
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019944489297996892,
      "loss": 0.0597,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.06911394000053406,
      "eval_runtime": 654.3952,
      "eval_samples_per_second": 11.244,
      "eval_steps_per_second": 2.812,
      "step": 1000
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019943374587080333,
      "loss": 0.0674,
      "step": 1010
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001994224882666228,
      "loss": 0.0528,
      "step": 1020
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001994111201799373,
      "loss": 0.0602,
      "step": 1030
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019939964162337946,
      "loss": 0.0568,
      "step": 1040
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019938805260970486,
      "loss": 0.0571,
      "step": 1050
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001993763531517916,
      "loss": 0.0598,
      "step": 1060
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019936454326264068,
      "loss": 0.0636,
      "step": 1070
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019935262295537568,
      "loss": 0.0602,
      "step": 1080
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019934059224324303,
      "loss": 0.0763,
      "step": 1090
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019932845113961172,
      "loss": 0.0538,
      "step": 1100
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.06597457081079483,
      "eval_runtime": 653.2731,
      "eval_samples_per_second": 11.263,
      "eval_steps_per_second": 2.817,
      "step": 1100
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001993161996579735,
      "loss": 0.0578,
      "step": 1110
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019930383781194272,
      "loss": 0.0561,
      "step": 1120
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019929136561525637,
      "loss": 0.0558,
      "step": 1130
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001992787830817741,
      "loss": 0.0669,
      "step": 1140
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019926609022547818,
      "loss": 0.0616,
      "step": 1150
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019925328706047344,
      "loss": 0.0637,
      "step": 1160
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019924037360098727,
      "loss": 0.052,
      "step": 1170
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001992273498613697,
      "loss": 0.0548,
      "step": 1180
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019921421585609325,
      "loss": 0.0545,
      "step": 1190
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000199200971599753,
      "loss": 0.0498,
      "step": 1200
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.06580409407615662,
      "eval_runtime": 655.2658,
      "eval_samples_per_second": 11.229,
      "eval_steps_per_second": 2.808,
      "step": 1200
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019918761710706649,
      "loss": 0.0537,
      "step": 1210
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019917415239287386,
      "loss": 0.0579,
      "step": 1220
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019916057747213758,
      "loss": 0.053,
      "step": 1230
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019914689235994278,
      "loss": 0.0562,
      "step": 1240
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019913309707149694,
      "loss": 0.0595,
      "step": 1250
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019911919162212988,
      "loss": 0.0515,
      "step": 1260
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000199105176027294,
      "loss": 0.053,
      "step": 1270
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019909105030256402,
      "loss": 0.065,
      "step": 1280
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.000199076814463637,
      "loss": 0.0574,
      "step": 1290
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001990624685263325,
      "loss": 0.0595,
      "step": 1300
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.06419895589351654,
      "eval_runtime": 662.3031,
      "eval_samples_per_second": 11.11,
      "eval_steps_per_second": 2.778,
      "step": 1300
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019904801250659223,
      "loss": 0.0667,
      "step": 1310
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019903344642048043,
      "loss": 0.0562,
      "step": 1320
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019901877028418347,
      "loss": 0.0604,
      "step": 1330
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019900398411401015,
      "loss": 0.0474,
      "step": 1340
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001989890879263915,
      "loss": 0.0444,
      "step": 1350
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001989740817378808,
      "loss": 0.0522,
      "step": 1360
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019895896556515352,
      "loss": 0.0543,
      "step": 1370
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019894373942500743,
      "loss": 0.0447,
      "step": 1380
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019892840333436245,
      "loss": 0.0454,
      "step": 1390
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019891295731026075,
      "loss": 0.0465,
      "step": 1400
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.0676579549908638,
      "eval_runtime": 660.0804,
      "eval_samples_per_second": 11.147,
      "eval_steps_per_second": 2.788,
      "step": 1400
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001988974013698665,
      "loss": 0.0533,
      "step": 1410
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019888173553046624,
      "loss": 0.0522,
      "step": 1420
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019886595980946842,
      "loss": 0.0592,
      "step": 1430
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019885007422440379,
      "loss": 0.0584,
      "step": 1440
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019883407879292498,
      "loss": 0.054,
      "step": 1450
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019881797353280695,
      "loss": 0.0442,
      "step": 1460
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019880175846194638,
      "loss": 0.0659,
      "step": 1470
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019878543359836223,
      "loss": 0.0483,
      "step": 1480
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001987689989601954,
      "loss": 0.0525,
      "step": 1490
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019875245456570873,
      "loss": 0.0533,
      "step": 1500
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.06512827426195145,
      "eval_runtime": 659.6413,
      "eval_samples_per_second": 11.155,
      "eval_steps_per_second": 2.789,
      "step": 1500
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019873580043328705,
      "loss": 0.0438,
      "step": 1510
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019871903658143719,
      "loss": 0.0542,
      "step": 1520
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001987021630287878,
      "loss": 0.0555,
      "step": 1530
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019868517979408954,
      "loss": 0.0561,
      "step": 1540
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019866808689621482,
      "loss": 0.0517,
      "step": 1550
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001986508843541581,
      "loss": 0.051,
      "step": 1560
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001986335721870355,
      "loss": 0.0525,
      "step": 1570
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001986161504140851,
      "loss": 0.0443,
      "step": 1580
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019859861905466668,
      "loss": 0.0542,
      "step": 1590
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019858097812826183,
      "loss": 0.0593,
      "step": 1600
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.06405352801084518,
      "eval_runtime": 658.5179,
      "eval_samples_per_second": 11.174,
      "eval_steps_per_second": 2.794,
      "step": 1600
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001985632276544739,
      "loss": 0.0527,
      "step": 1610
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000198545367653028,
      "loss": 0.0422,
      "step": 1620
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019852739814377087,
      "loss": 0.0533,
      "step": 1630
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019850931914667107,
      "loss": 0.0507,
      "step": 1640
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019849113068181869,
      "loss": 0.0581,
      "step": 1650
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019847283276942554,
      "loss": 0.0583,
      "step": 1660
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019845442542982506,
      "loss": 0.0562,
      "step": 1670
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019843590868347225,
      "loss": 0.0499,
      "step": 1680
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019841728255094374,
      "loss": 0.0543,
      "step": 1690
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019839854705293764,
      "loss": 0.055,
      "step": 1700
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.0653354823589325,
      "eval_runtime": 674.0253,
      "eval_samples_per_second": 10.917,
      "eval_steps_per_second": 2.73,
      "step": 1700
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019837970221027365,
      "loss": 0.0422,
      "step": 1710
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019836074804389296,
      "loss": 0.0586,
      "step": 1720
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019834168457485824,
      "loss": 0.0408,
      "step": 1730
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019832251182435367,
      "loss": 0.0587,
      "step": 1740
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019830322981368478,
      "loss": 0.0498,
      "step": 1750
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001982838385642786,
      "loss": 0.0475,
      "step": 1760
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019826433809768345,
      "loss": 0.0571,
      "step": 1770
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019824472843556914,
      "loss": 0.0512,
      "step": 1780
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019822500959972673,
      "loss": 0.0477,
      "step": 1790
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019820518161206864,
      "loss": 0.0546,
      "step": 1800
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.06339588016271591,
      "eval_runtime": 1055.2159,
      "eval_samples_per_second": 6.973,
      "eval_steps_per_second": 1.744,
      "step": 1800
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019818524449462862,
      "loss": 0.0463,
      "step": 1810
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019816519826956156,
      "loss": 0.0571,
      "step": 1820
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019814504295914372,
      "loss": 0.0516,
      "step": 1830
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019812477858577258,
      "loss": 0.0511,
      "step": 1840
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019810440517196672,
      "loss": 0.0471,
      "step": 1850
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019808392274036596,
      "loss": 0.0497,
      "step": 1860
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019806333131373126,
      "loss": 0.0608,
      "step": 1870
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019804263091494466,
      "loss": 0.0431,
      "step": 1880
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019802182156700938,
      "loss": 0.0558,
      "step": 1890
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019800090329304956,
      "loss": 0.0524,
      "step": 1900
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.06149962544441223,
      "eval_runtime": 1053.8269,
      "eval_samples_per_second": 6.982,
      "eval_steps_per_second": 1.746,
      "step": 1900
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019797987611631059,
      "loss": 0.0528,
      "step": 1910
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019795874006015872,
      "loss": 0.0481,
      "step": 1920
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001979374951480812,
      "loss": 0.0491,
      "step": 1930
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019791614140368633,
      "loss": 0.049,
      "step": 1940
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019789467885070327,
      "loss": 0.0444,
      "step": 1950
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019787310751298213,
      "loss": 0.0466,
      "step": 1960
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019785142741449393,
      "loss": 0.0498,
      "step": 1970
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019782963857933048,
      "loss": 0.0475,
      "step": 1980
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019780774103170446,
      "loss": 0.047,
      "step": 1990
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019778573479594942,
      "loss": 0.0432,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.06317023187875748,
      "eval_runtime": 1054.8968,
      "eval_samples_per_second": 6.975,
      "eval_steps_per_second": 1.744,
      "step": 2000
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019776361989651956,
      "loss": 0.0546,
      "step": 2010
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001977413963579899,
      "loss": 0.043,
      "step": 2020
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019771906420505624,
      "loss": 0.0579,
      "step": 2030
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019769662346253493,
      "loss": 0.057,
      "step": 2040
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019767407415536317,
      "loss": 0.0476,
      "step": 2050
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019765141630859865,
      "loss": 0.0418,
      "step": 2060
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019762864994741976,
      "loss": 0.0497,
      "step": 2070
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019760577509712546,
      "loss": 0.0493,
      "step": 2080
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019758279178313525,
      "loss": 0.0512,
      "step": 2090
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019755970003098916,
      "loss": 0.0631,
      "step": 2100
    },
    {
      "epoch": 0.42,
      "eval_loss": 0.06188439950346947,
      "eval_runtime": 1062.072,
      "eval_samples_per_second": 6.928,
      "eval_steps_per_second": 1.732,
      "step": 2100
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019753649986634772,
      "loss": 0.0543,
      "step": 2110
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019751319131499194,
      "loss": 0.0405,
      "step": 2120
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019748977440282333,
      "loss": 0.0564,
      "step": 2130
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001974662491558637,
      "loss": 0.0396,
      "step": 2140
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019744261560025533,
      "loss": 0.039,
      "step": 2150
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019741887376226083,
      "loss": 0.0348,
      "step": 2160
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019739502366826313,
      "loss": 0.0533,
      "step": 2170
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001973710653447655,
      "loss": 0.0417,
      "step": 2180
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001973469988183914,
      "loss": 0.0502,
      "step": 2190
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019732282411588463,
      "loss": 0.0519,
      "step": 2200
    },
    {
      "epoch": 0.44,
      "eval_loss": 0.05978045240044594,
      "eval_runtime": 1059.6808,
      "eval_samples_per_second": 6.944,
      "eval_steps_per_second": 1.736,
      "step": 2200
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019729854126410913,
      "loss": 0.0416,
      "step": 2210
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019727415029004906,
      "loss": 0.0586,
      "step": 2220
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019724965122080868,
      "loss": 0.0535,
      "step": 2230
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001972250440836124,
      "loss": 0.0499,
      "step": 2240
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019720032890580474,
      "loss": 0.0484,
      "step": 2250
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019717550571485024,
      "loss": 0.044,
      "step": 2260
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001971505745383335,
      "loss": 0.0496,
      "step": 2270
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019712553540395908,
      "loss": 0.05,
      "step": 2280
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001971003883395516,
      "loss": 0.0387,
      "step": 2290
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019707513337305547,
      "loss": 0.0397,
      "step": 2300
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.060694798827171326,
      "eval_runtime": 1049.503,
      "eval_samples_per_second": 7.011,
      "eval_steps_per_second": 1.753,
      "step": 2300
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001970497705325351,
      "loss": 0.0475,
      "step": 2310
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019702429984617484,
      "loss": 0.0405,
      "step": 2320
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019699872134227867,
      "loss": 0.0447,
      "step": 2330
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019697303504927061,
      "loss": 0.0608,
      "step": 2340
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019694724099569434,
      "loss": 0.0478,
      "step": 2350
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019692133921021332,
      "loss": 0.0432,
      "step": 2360
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019689532972161068,
      "loss": 0.0492,
      "step": 2370
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019686921255878932,
      "loss": 0.0441,
      "step": 2380
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001968429877507717,
      "loss": 0.0603,
      "step": 2390
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019681665532669996,
      "loss": 0.0467,
      "step": 2400
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.06163698434829712,
      "eval_runtime": 1049.1311,
      "eval_samples_per_second": 7.013,
      "eval_steps_per_second": 1.754,
      "step": 2400
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019679021531583584,
      "loss": 0.0406,
      "step": 2410
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019676366774756056,
      "loss": 0.0484,
      "step": 2420
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019673701265137495,
      "loss": 0.0402,
      "step": 2430
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019671025005689926,
      "loss": 0.0541,
      "step": 2440
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019668337999387324,
      "loss": 0.0438,
      "step": 2450
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019665640249215605,
      "loss": 0.045,
      "step": 2460
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001966293175817262,
      "loss": 0.0456,
      "step": 2470
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019660212529268168,
      "loss": 0.0454,
      "step": 2480
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019657482565523963,
      "loss": 0.0464,
      "step": 2490
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019654741869973663,
      "loss": 0.049,
      "step": 2500
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.06358644366264343,
      "eval_runtime": 1050.5846,
      "eval_samples_per_second": 7.004,
      "eval_steps_per_second": 1.751,
      "step": 2500
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019651990445662841,
      "loss": 0.053,
      "step": 2510
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019649228295649004,
      "loss": 0.05,
      "step": 2520
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019646455423001565,
      "loss": 0.0423,
      "step": 2530
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001964367183080186,
      "loss": 0.0556,
      "step": 2540
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019640877522143134,
      "loss": 0.0431,
      "step": 2550
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001963807250013054,
      "loss": 0.0377,
      "step": 2560
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019635256767881144,
      "loss": 0.0432,
      "step": 2570
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019632430328523902,
      "loss": 0.0562,
      "step": 2580
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001962959318519968,
      "loss": 0.04,
      "step": 2590
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019626745341061225,
      "loss": 0.0488,
      "step": 2600
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.06040577217936516,
      "eval_runtime": 1049.7742,
      "eval_samples_per_second": 7.009,
      "eval_steps_per_second": 1.753,
      "step": 2600
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001962388679927319,
      "loss": 0.0371,
      "step": 2610
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.000196210175630121,
      "loss": 0.0412,
      "step": 2620
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019618137635466382,
      "loss": 0.0432,
      "step": 2630
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019615247019836327,
      "loss": 0.0501,
      "step": 2640
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019612345719334116,
      "loss": 0.0422,
      "step": 2650
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019609433737183791,
      "loss": 0.0511,
      "step": 2660
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019606511076621276,
      "loss": 0.0372,
      "step": 2670
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019603577740894354,
      "loss": 0.0541,
      "step": 2680
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001960063373326267,
      "loss": 0.0472,
      "step": 2690
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019597679056997737,
      "loss": 0.0449,
      "step": 2700
    },
    {
      "epoch": 0.54,
      "eval_loss": 0.05977201834321022,
      "eval_runtime": 1052.2024,
      "eval_samples_per_second": 6.993,
      "eval_steps_per_second": 1.749,
      "step": 2700
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019594713715382915,
      "loss": 0.0517,
      "step": 2710
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019591737711713414,
      "loss": 0.0476,
      "step": 2720
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019588751049296298,
      "loss": 0.048,
      "step": 2730
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019585753731450478,
      "loss": 0.0342,
      "step": 2740
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019582745761506697,
      "loss": 0.0508,
      "step": 2750
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019579727142807535,
      "loss": 0.0407,
      "step": 2760
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001957669787870742,
      "loss": 0.0358,
      "step": 2770
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019573657972572593,
      "loss": 0.0506,
      "step": 2780
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019570607427781128,
      "loss": 0.0476,
      "step": 2790
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001956754624772292,
      "loss": 0.0438,
      "step": 2800
    },
    {
      "epoch": 0.56,
      "eval_loss": 0.061029911041259766,
      "eval_runtime": 1062.1739,
      "eval_samples_per_second": 6.927,
      "eval_steps_per_second": 1.732,
      "step": 2800
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001956447443579968,
      "loss": 0.0481,
      "step": 2810
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019561391995424941,
      "loss": 0.0515,
      "step": 2820
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019558298930024044,
      "loss": 0.0491,
      "step": 2830
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001955519524303413,
      "loss": 0.0367,
      "step": 2840
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001955208093790415,
      "loss": 0.0431,
      "step": 2850
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001954895601809485,
      "loss": 0.0436,
      "step": 2860
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001954582048707878,
      "loss": 0.0442,
      "step": 2870
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019542674348340267,
      "loss": 0.0532,
      "step": 2880
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019539517605375446,
      "loss": 0.0441,
      "step": 2890
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019536350261692214,
      "loss": 0.036,
      "step": 2900
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.06326338648796082,
      "eval_runtime": 1061.8457,
      "eval_samples_per_second": 6.929,
      "eval_steps_per_second": 1.733,
      "step": 2900
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019533172320810265,
      "loss": 0.0486,
      "step": 2910
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019529983786261058,
      "loss": 0.0568,
      "step": 2920
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019526784661587829,
      "loss": 0.0434,
      "step": 2930
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001952357495034558,
      "loss": 0.0541,
      "step": 2940
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019520354656101085,
      "loss": 0.0395,
      "step": 2950
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019517123782432868,
      "loss": 0.0458,
      "step": 2960
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019513882332931212,
      "loss": 0.0395,
      "step": 2970
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019510630311198157,
      "loss": 0.0442,
      "step": 2980
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019507367720847488,
      "loss": 0.041,
      "step": 2990
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019504094565504733,
      "loss": 0.0464,
      "step": 3000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.06028781458735466,
      "eval_runtime": 1035.9718,
      "eval_samples_per_second": 7.103,
      "eval_steps_per_second": 1.776,
      "step": 3000
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00019500810848807162,
      "loss": 0.0396,
      "step": 3010
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00019497516574403778,
      "loss": 0.0491,
      "step": 3020
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00019494211745955324,
      "loss": 0.0585,
      "step": 3030
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00019490896367134266,
      "loss": 0.0484,
      "step": 3040
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00019487570441624791,
      "loss": 0.0507,
      "step": 3050
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001948423397312281,
      "loss": 0.0467,
      "step": 3060
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001948088696533595,
      "loss": 0.0444,
      "step": 3070
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00019477529421983546,
      "loss": 0.0498,
      "step": 3080
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001947416134679665,
      "loss": 0.0373,
      "step": 3090
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00019470782743518002,
      "loss": 0.0437,
      "step": 3100
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.06118550896644592,
      "eval_runtime": 1055.3909,
      "eval_samples_per_second": 6.972,
      "eval_steps_per_second": 1.743,
      "step": 3100
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00019467393615902055,
      "loss": 0.0391,
      "step": 3110
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001946399396771495,
      "loss": 0.0456,
      "step": 3120
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00019460583802734523,
      "loss": 0.0368,
      "step": 3130
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001945716312475029,
      "loss": 0.0451,
      "step": 3140
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001945373193756346,
      "loss": 0.046,
      "step": 3150
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00019450290244986914,
      "loss": 0.0364,
      "step": 3160
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00019446838050845205,
      "loss": 0.0576,
      "step": 3170
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00019443375358974555,
      "loss": 0.0421,
      "step": 3180
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00019439902173222859,
      "loss": 0.0528,
      "step": 3190
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001943641849744967,
      "loss": 0.0389,
      "step": 3200
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.06049993634223938,
      "eval_runtime": 1052.016,
      "eval_samples_per_second": 6.994,
      "eval_steps_per_second": 1.749,
      "step": 3200
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00019432924335526194,
      "loss": 0.0432,
      "step": 3210
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00019429419691335297,
      "loss": 0.0339,
      "step": 3220
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00019425904568771483,
      "loss": 0.0434,
      "step": 3230
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00019422378971740907,
      "loss": 0.0472,
      "step": 3240
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00019418842904161368,
      "loss": 0.0394,
      "step": 3250
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019415296369962288,
      "loss": 0.0404,
      "step": 3260
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019411739373084732,
      "loss": 0.0352,
      "step": 3270
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019408171917481386,
      "loss": 0.0364,
      "step": 3280
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019404594007116555,
      "loss": 0.0433,
      "step": 3290
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00019401005645966167,
      "loss": 0.0377,
      "step": 3300
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.06322025507688522,
      "eval_runtime": 1050.1226,
      "eval_samples_per_second": 7.007,
      "eval_steps_per_second": 1.752,
      "step": 3300
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00019397406838017766,
      "loss": 0.042,
      "step": 3310
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00019393797587270497,
      "loss": 0.0326,
      "step": 3320
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00019390177897735114,
      "loss": 0.0443,
      "step": 3330
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00019386547773433965,
      "loss": 0.0545,
      "step": 3340
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00019382907218401006,
      "loss": 0.0422,
      "step": 3350
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00019379256236681775,
      "loss": 0.0393,
      "step": 3360
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.000193755948323334,
      "loss": 0.0516,
      "step": 3370
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00019371923009424587,
      "loss": 0.0499,
      "step": 3380
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001936824077203562,
      "loss": 0.0491,
      "step": 3390
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001936454812425836,
      "loss": 0.0382,
      "step": 3400
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.06261178106069565,
      "eval_runtime": 1054.0723,
      "eval_samples_per_second": 6.981,
      "eval_steps_per_second": 1.746,
      "step": 3400
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00019360845070196236,
      "loss": 0.0431,
      "step": 3410
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001935713161396424,
      "loss": 0.0498,
      "step": 3420
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001935340775968892,
      "loss": 0.0457,
      "step": 3430
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00019349673511508383,
      "loss": 0.0362,
      "step": 3440
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00019345928873572282,
      "loss": 0.0432,
      "step": 3450
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00019342173850041822,
      "loss": 0.0365,
      "step": 3460
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00019338408445089745,
      "loss": 0.0358,
      "step": 3470
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001933463266290033,
      "loss": 0.039,
      "step": 3480
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00019330846507669382,
      "loss": 0.0412,
      "step": 3490
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00019327049983604245,
      "loss": 0.051,
      "step": 3500
    },
    {
      "epoch": 0.7,
      "eval_loss": 0.061244383454322815,
      "eval_runtime": 1045.5529,
      "eval_samples_per_second": 7.037,
      "eval_steps_per_second": 1.76,
      "step": 3500
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00019323243094923772,
      "loss": 0.0387,
      "step": 3510
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00019319425845858341,
      "loss": 0.0473,
      "step": 3520
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00019315598240649847,
      "loss": 0.037,
      "step": 3530
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001931176028355168,
      "loss": 0.0435,
      "step": 3540
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00019307911978828747,
      "loss": 0.0364,
      "step": 3550
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001930405333075745,
      "loss": 0.0424,
      "step": 3560
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019300184343625678,
      "loss": 0.0348,
      "step": 3570
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019296305021732817,
      "loss": 0.0395,
      "step": 3580
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019292415369389734,
      "loss": 0.0486,
      "step": 3590
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00019288515390918776,
      "loss": 0.047,
      "step": 3600
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.06326541304588318,
      "eval_runtime": 1046.8433,
      "eval_samples_per_second": 7.029,
      "eval_steps_per_second": 1.758,
      "step": 3600
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019284605090653766,
      "loss": 0.0403,
      "step": 3610
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019280684472939994,
      "loss": 0.0381,
      "step": 3620
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019276753542134224,
      "loss": 0.0332,
      "step": 3630
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019272812302604665,
      "loss": 0.0435,
      "step": 3640
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00019268860758730997,
      "loss": 0.0512,
      "step": 3650
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00019264898914904342,
      "loss": 0.0403,
      "step": 3660
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00019260926775527265,
      "loss": 0.0336,
      "step": 3670
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00019256944345013785,
      "loss": 0.0425,
      "step": 3680
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00019252951627789344,
      "loss": 0.0431,
      "step": 3690
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00019248948628290818,
      "loss": 0.0451,
      "step": 3700
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.06083804368972778,
      "eval_runtime": 1049.5583,
      "eval_samples_per_second": 7.011,
      "eval_steps_per_second": 1.753,
      "step": 3700
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019244935350966514,
      "loss": 0.032,
      "step": 3710
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019240911800276153,
      "loss": 0.0378,
      "step": 3720
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001923687798069088,
      "loss": 0.044,
      "step": 3730
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019232833896693242,
      "loss": 0.0363,
      "step": 3740
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00019228779552777202,
      "loss": 0.0412,
      "step": 3750
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001922471495344812,
      "loss": 0.0376,
      "step": 3760
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001922064010322275,
      "loss": 0.0451,
      "step": 3770
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00019216555006629237,
      "loss": 0.0443,
      "step": 3780
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001921245966820712,
      "loss": 0.0394,
      "step": 3790
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00019208354092507305,
      "loss": 0.0472,
      "step": 3800
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.06302132457494736,
      "eval_runtime": 1043.9496,
      "eval_samples_per_second": 7.048,
      "eval_steps_per_second": 1.763,
      "step": 3800
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00019204238284092093,
      "loss": 0.0418,
      "step": 3810
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00019200112247535141,
      "loss": 0.0416,
      "step": 3820
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00019195975987421472,
      "loss": 0.0418,
      "step": 3830
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00019191829508347481,
      "loss": 0.0394,
      "step": 3840
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00019187672814920912,
      "loss": 0.0405,
      "step": 3850
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00019183505911760855,
      "loss": 0.0306,
      "step": 3860
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00019179328803497754,
      "loss": 0.0443,
      "step": 3870
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001917514149477339,
      "loss": 0.035,
      "step": 3880
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00019170943990240877,
      "loss": 0.0371,
      "step": 3890
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001916673629456466,
      "loss": 0.0394,
      "step": 3900
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.06118810176849365,
      "eval_runtime": 1046.2783,
      "eval_samples_per_second": 7.033,
      "eval_steps_per_second": 1.759,
      "step": 3900
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00019162518412420512,
      "loss": 0.0337,
      "step": 3910
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00019158290348495524,
      "loss": 0.0433,
      "step": 3920
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.000191540521074881,
      "loss": 0.0402,
      "step": 3930
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001914980369410795,
      "loss": 0.0326,
      "step": 3940
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00019145545113076096,
      "loss": 0.0501,
      "step": 3950
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00019141276369124855,
      "loss": 0.0448,
      "step": 3960
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00019136997466997837,
      "loss": 0.0403,
      "step": 3970
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00019132708411449936,
      "loss": 0.0427,
      "step": 3980
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001912840920724734,
      "loss": 0.0322,
      "step": 3990
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00019124099859167503,
      "loss": 0.0471,
      "step": 4000
    },
    {
      "epoch": 0.81,
      "eval_loss": 0.06139204651117325,
      "eval_runtime": 1057.3652,
      "eval_samples_per_second": 6.959,
      "eval_steps_per_second": 1.74,
      "step": 4000
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00019119780371999162,
      "loss": 0.0306,
      "step": 4010
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00019115450750542304,
      "loss": 0.0352,
      "step": 4020
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.000191111109996082,
      "loss": 0.0351,
      "step": 4030
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00019106761124019364,
      "loss": 0.0295,
      "step": 4040
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00019102401128609557,
      "loss": 0.0379,
      "step": 4050
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00019098031018223796,
      "loss": 0.048,
      "step": 4060
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00019093650797718338,
      "loss": 0.0454,
      "step": 4070
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00019089260471960663,
      "loss": 0.0463,
      "step": 4080
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001908486004582949,
      "loss": 0.0447,
      "step": 4090
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00019080449524214762,
      "loss": 0.031,
      "step": 4100
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.06018984317779541,
      "eval_runtime": 1049.9509,
      "eval_samples_per_second": 7.008,
      "eval_steps_per_second": 1.752,
      "step": 4100
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00019076028912017642,
      "loss": 0.0292,
      "step": 4110
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00019071598214150494,
      "loss": 0.0252,
      "step": 4120
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00019067157435536904,
      "loss": 0.0382,
      "step": 4130
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00019062706581111653,
      "loss": 0.0303,
      "step": 4140
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001905824565582072,
      "loss": 0.0419,
      "step": 4150
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00019053774664621272,
      "loss": 0.0358,
      "step": 4160
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00019049293612481675,
      "loss": 0.037,
      "step": 4170
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00019044802504381453,
      "loss": 0.0306,
      "step": 4180
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00019040301345311326,
      "loss": 0.0431,
      "step": 4190
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001903579014027317,
      "loss": 0.0405,
      "step": 4200
    },
    {
      "epoch": 0.85,
      "eval_loss": 0.05838855355978012,
      "eval_runtime": 1051.4215,
      "eval_samples_per_second": 6.998,
      "eval_steps_per_second": 1.75,
      "step": 4200
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00019031268894280023,
      "loss": 0.0334,
      "step": 4210
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00019026737612356094,
      "loss": 0.0382,
      "step": 4220
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00019022196299536733,
      "loss": 0.0371,
      "step": 4230
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00019017644960868445,
      "loss": 0.0353,
      "step": 4240
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00019013083601408863,
      "loss": 0.0368,
      "step": 4250
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001900851222622677,
      "loss": 0.0427,
      "step": 4260
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00019003930840402072,
      "loss": 0.031,
      "step": 4270
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00018999339449025796,
      "loss": 0.0377,
      "step": 4280
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00018994738057200099,
      "loss": 0.0373,
      "step": 4290
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001899012667003824,
      "loss": 0.0421,
      "step": 4300
    },
    {
      "epoch": 0.87,
      "eval_loss": 0.05914044752717018,
      "eval_runtime": 1056.4599,
      "eval_samples_per_second": 6.965,
      "eval_steps_per_second": 1.742,
      "step": 4300
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018985505292664587,
      "loss": 0.0344,
      "step": 4310
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018980873930214614,
      "loss": 0.0355,
      "step": 4320
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018976232587834886,
      "loss": 0.0416,
      "step": 4330
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018971581270683062,
      "loss": 0.0395,
      "step": 4340
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00018966919983927886,
      "loss": 0.0366,
      "step": 4350
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00018962248732749175,
      "loss": 0.0368,
      "step": 4360
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001895756752233782,
      "loss": 0.0362,
      "step": 4370
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0001895287635789579,
      "loss": 0.0395,
      "step": 4380
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00018948175244636097,
      "loss": 0.0462,
      "step": 4390
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00018943464187782828,
      "loss": 0.0454,
      "step": 4400
    },
    {
      "epoch": 0.89,
      "eval_loss": 0.05928103253245354,
      "eval_runtime": 1029.6162,
      "eval_samples_per_second": 7.146,
      "eval_steps_per_second": 1.787,
      "step": 4400
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.000189387431925711,
      "loss": 0.0326,
      "step": 4410
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001893401226424709,
      "loss": 0.031,
      "step": 4420
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00018929271408068011,
      "loss": 0.0408,
      "step": 4430
    },
    {
      "epoch": 0.89,
      "step": 4431,
      "total_flos": 1.532304306097619e+18,
      "train_loss": 9.867495303766314e-06,
      "train_runtime": 4.4906,
      "train_samples_per_second": 15784.097,
      "train_steps_per_second": 986.506
    }
  ],
  "logging_steps": 10,
  "max_steps": 4430,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 1.532304306097619e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}