{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.75609756097561,
  "global_step": 2800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 0.00019930313588850174,
      "loss": 3.8899,
      "step": 10
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019860627177700348,
      "loss": 3.223,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019790940766550524,
      "loss": 3.1045,
      "step": 30
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019721254355400697,
      "loss": 2.847,
      "step": 40
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019651567944250874,
      "loss": 2.9753,
      "step": 50
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019581881533101044,
      "loss": 2.9102,
      "step": 60
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001951219512195122,
      "loss": 2.8334,
      "step": 70
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019442508710801394,
      "loss": 2.794,
      "step": 80
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001937282229965157,
      "loss": 2.7356,
      "step": 90
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001930313588850174,
      "loss": 2.6414,
      "step": 100
    },
    {
      "epoch": 0.35,
      "eval_accuracy": 0.5204419889502763,
      "eval_loss": 2.669311046600342,
      "eval_runtime": 293.1372,
      "eval_samples_per_second": 27.786,
      "eval_steps_per_second": 3.476,
      "step": 100
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019233449477351917,
      "loss": 2.6534,
      "step": 110
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001916376306620209,
      "loss": 2.5765,
      "step": 120
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019094076655052267,
      "loss": 2.5439,
      "step": 130
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001902439024390244,
      "loss": 2.5482,
      "step": 140
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00018954703832752614,
      "loss": 2.5077,
      "step": 150
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001888501742160279,
      "loss": 2.5218,
      "step": 160
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00018815331010452963,
      "loss": 2.3255,
      "step": 170
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00018745644599303137,
      "loss": 2.3874,
      "step": 180
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001867595818815331,
      "loss": 2.3304,
      "step": 190
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00018606271777003486,
      "loss": 2.2286,
      "step": 200
    },
    {
      "epoch": 0.7,
      "eval_accuracy": 0.5642725598526703,
      "eval_loss": 2.2449886798858643,
      "eval_runtime": 293.6028,
      "eval_samples_per_second": 27.742,
      "eval_steps_per_second": 3.471,
      "step": 200
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001853658536585366,
      "loss": 2.1984,
      "step": 210
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00018466898954703833,
      "loss": 2.2406,
      "step": 220
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00018397212543554007,
      "loss": 2.1315,
      "step": 230
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00018327526132404183,
      "loss": 2.0743,
      "step": 240
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00018257839721254356,
      "loss": 2.0483,
      "step": 250
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001818815331010453,
      "loss": 2.0747,
      "step": 260
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00018118466898954703,
      "loss": 2.0846,
      "step": 270
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001804878048780488,
      "loss": 1.9392,
      "step": 280
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017979094076655053,
      "loss": 1.9443,
      "step": 290
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00017909407665505226,
      "loss": 1.871,
      "step": 300
    },
    {
      "epoch": 1.05,
      "eval_accuracy": 0.6078575813382443,
      "eval_loss": 1.9149649143218994,
      "eval_runtime": 301.3751,
      "eval_samples_per_second": 27.026,
      "eval_steps_per_second": 3.381,
      "step": 300
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00017839721254355403,
      "loss": 1.9009,
      "step": 310
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00017770034843205576,
      "loss": 1.8669,
      "step": 320
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001770034843205575,
      "loss": 1.8627,
      "step": 330
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00017630662020905923,
      "loss": 1.8475,
      "step": 340
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.000175609756097561,
      "loss": 1.7538,
      "step": 350
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00017491289198606273,
      "loss": 1.7642,
      "step": 360
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00017421602787456446,
      "loss": 1.7665,
      "step": 370
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001735191637630662,
      "loss": 1.6783,
      "step": 380
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00017282229965156796,
      "loss": 1.6583,
      "step": 390
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001721254355400697,
      "loss": 1.7339,
      "step": 400
    },
    {
      "epoch": 1.39,
      "eval_accuracy": 0.6306936771025169,
      "eval_loss": 1.6812902688980103,
      "eval_runtime": 292.2248,
      "eval_samples_per_second": 27.872,
      "eval_steps_per_second": 3.487,
      "step": 400
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00017142857142857143,
      "loss": 1.6163,
      "step": 410
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001707317073170732,
      "loss": 1.5708,
      "step": 420
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00017003484320557492,
      "loss": 1.5696,
      "step": 430
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00016933797909407668,
      "loss": 1.6017,
      "step": 440
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001686411149825784,
      "loss": 1.5735,
      "step": 450
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00016794425087108015,
      "loss": 1.5265,
      "step": 460
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001672473867595819,
      "loss": 1.5508,
      "step": 470
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00016655052264808365,
      "loss": 1.4604,
      "step": 480
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00016585365853658536,
      "loss": 1.5646,
      "step": 490
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00016515679442508712,
      "loss": 1.484,
      "step": 500
    },
    {
      "epoch": 1.74,
      "eval_accuracy": 0.6531614487415592,
      "eval_loss": 1.5198744535446167,
      "eval_runtime": 291.9876,
      "eval_samples_per_second": 27.895,
      "eval_steps_per_second": 3.49,
      "step": 500
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00016445993031358885,
      "loss": 1.5005,
      "step": 510
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00016376306620209061,
      "loss": 1.4561,
      "step": 520
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00016306620209059232,
      "loss": 1.4335,
      "step": 530
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00016236933797909408,
      "loss": 1.4252,
      "step": 540
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00016167247386759582,
      "loss": 1.4586,
      "step": 550
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00016097560975609758,
      "loss": 1.4068,
      "step": 560
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00016027874564459931,
      "loss": 1.4227,
      "step": 570
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.00015958188153310105,
      "loss": 1.3449,
      "step": 580
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0001588850174216028,
      "loss": 1.308,
      "step": 590
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.00015818815331010454,
      "loss": 1.3087,
      "step": 600
    },
    {
      "epoch": 2.09,
      "eval_accuracy": 0.6647022713321056,
      "eval_loss": 1.4196171760559082,
      "eval_runtime": 293.0401,
      "eval_samples_per_second": 27.795,
      "eval_steps_per_second": 3.477,
      "step": 600
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.00015749128919860628,
      "loss": 1.2248,
      "step": 610
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00015679442508710801,
      "loss": 1.2168,
      "step": 620
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.00015609756097560978,
      "loss": 1.2282,
      "step": 630
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.00015540069686411148,
      "loss": 1.2161,
      "step": 640
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.00015470383275261324,
      "loss": 1.1724,
      "step": 650
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.00015400696864111498,
      "loss": 1.2168,
      "step": 660
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.00015331010452961674,
      "loss": 1.1928,
      "step": 670
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.00015261324041811848,
      "loss": 1.1841,
      "step": 680
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0001519163763066202,
      "loss": 1.1963,
      "step": 690
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.00015121951219512197,
      "loss": 1.1799,
      "step": 700
    },
    {
      "epoch": 2.44,
      "eval_accuracy": 0.6929404542664211,
      "eval_loss": 1.2876527309417725,
      "eval_runtime": 291.746,
      "eval_samples_per_second": 27.918,
      "eval_steps_per_second": 3.493,
      "step": 700
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0001505226480836237,
      "loss": 1.1571,
      "step": 710
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.00014982578397212544,
      "loss": 1.1253,
      "step": 720
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.00014912891986062718,
      "loss": 1.1619,
      "step": 730
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.00014843205574912894,
      "loss": 1.1534,
      "step": 740
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.00014773519163763067,
      "loss": 1.1443,
      "step": 750
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0001470383275261324,
      "loss": 1.1674,
      "step": 760
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.00014634146341463414,
      "loss": 1.2101,
      "step": 770
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.0001456445993031359,
      "loss": 1.156,
      "step": 780
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.00014494773519163764,
      "loss": 1.1294,
      "step": 790
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.00014425087108013937,
      "loss": 1.1236,
      "step": 800
    },
    {
      "epoch": 2.79,
      "eval_accuracy": 0.7019030079803561,
      "eval_loss": 1.2286723852157593,
      "eval_runtime": 294.5533,
      "eval_samples_per_second": 27.652,
      "eval_steps_per_second": 3.459,
      "step": 800
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0001435540069686411,
      "loss": 1.1472,
      "step": 810
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.00014285714285714287,
      "loss": 1.1352,
      "step": 820
    },
    {
      "epoch": 2.89,
      "learning_rate": 0.0001421602787456446,
      "loss": 1.1197,
      "step": 830
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.00014146341463414634,
      "loss": 1.0767,
      "step": 840
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.0001407665505226481,
      "loss": 1.0855,
      "step": 850
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00014006968641114983,
      "loss": 1.12,
      "step": 860
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0001393728222996516,
      "loss": 1.0387,
      "step": 870
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0001386759581881533,
      "loss": 0.9932,
      "step": 880
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.00013797909407665506,
      "loss": 0.9766,
      "step": 890
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0001372822299651568,
      "loss": 0.9447,
      "step": 900
    },
    {
      "epoch": 3.14,
      "eval_accuracy": 0.7130755064456722,
      "eval_loss": 1.1540069580078125,
      "eval_runtime": 293.0443,
      "eval_samples_per_second": 27.794,
      "eval_steps_per_second": 3.477,
      "step": 900
    },
    {
      "epoch": 3.17,
      "learning_rate": 0.00013658536585365856,
      "loss": 0.9596,
      "step": 910
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.00013588850174216027,
      "loss": 0.9064,
      "step": 920
    },
    {
      "epoch": 3.24,
      "learning_rate": 0.00013519163763066203,
      "loss": 0.9208,
      "step": 930
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.00013449477351916376,
      "loss": 0.8917,
      "step": 940
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.00013379790940766553,
      "loss": 0.9753,
      "step": 950
    },
    {
      "epoch": 3.34,
      "learning_rate": 0.00013310104529616726,
      "loss": 0.947,
      "step": 960
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.000132404181184669,
      "loss": 0.9483,
      "step": 970
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.00013170731707317076,
      "loss": 0.92,
      "step": 980
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.00013101045296167246,
      "loss": 0.9128,
      "step": 990
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.00013031358885017423,
      "loss": 0.9463,
      "step": 1000
    },
    {
      "epoch": 3.48,
      "eval_accuracy": 0.7253529772866789,
      "eval_loss": 1.0846450328826904,
      "eval_runtime": 292.891,
      "eval_samples_per_second": 27.809,
      "eval_steps_per_second": 3.479,
      "step": 1000
    },
    {
      "epoch": 3.52,
      "learning_rate": 0.00012961672473867596,
      "loss": 0.916,
      "step": 1010
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.00012891986062717772,
      "loss": 0.9094,
      "step": 1020
    },
    {
      "epoch": 3.59,
      "learning_rate": 0.00012822299651567943,
      "loss": 0.9522,
      "step": 1030
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.0001275261324041812,
      "loss": 0.9032,
      "step": 1040
    },
    {
      "epoch": 3.66,
      "learning_rate": 0.00012682926829268293,
      "loss": 0.9051,
      "step": 1050
    },
    {
      "epoch": 3.69,
      "learning_rate": 0.0001261324041811847,
      "loss": 0.9214,
      "step": 1060
    },
    {
      "epoch": 3.73,
      "learning_rate": 0.0001254355400696864,
      "loss": 0.8808,
      "step": 1070
    },
    {
      "epoch": 3.76,
      "learning_rate": 0.00012473867595818816,
      "loss": 0.8724,
      "step": 1080
    },
    {
      "epoch": 3.8,
      "learning_rate": 0.0001240418118466899,
      "loss": 0.8951,
      "step": 1090
    },
    {
      "epoch": 3.83,
      "learning_rate": 0.00012334494773519165,
      "loss": 0.9383,
      "step": 1100
    },
    {
      "epoch": 3.83,
      "eval_accuracy": 0.7306322897483118,
      "eval_loss": 1.059184193611145,
      "eval_runtime": 293.0858,
      "eval_samples_per_second": 27.791,
      "eval_steps_per_second": 3.477,
      "step": 1100
    },
    {
      "epoch": 3.87,
      "learning_rate": 0.0001226480836236934,
      "loss": 0.8675,
      "step": 1110
    },
    {
      "epoch": 3.9,
      "learning_rate": 0.00012195121951219512,
      "loss": 0.8616,
      "step": 1120
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.00012125435540069687,
      "loss": 0.8807,
      "step": 1130
    },
    {
      "epoch": 3.97,
      "learning_rate": 0.00012055749128919862,
      "loss": 0.8594,
      "step": 1140
    },
    {
      "epoch": 4.01,
      "learning_rate": 0.00011986062717770035,
      "loss": 0.9073,
      "step": 1150
    },
    {
      "epoch": 4.04,
      "learning_rate": 0.0001191637630662021,
      "loss": 0.7567,
      "step": 1160
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.00011846689895470385,
      "loss": 0.7788,
      "step": 1170
    },
    {
      "epoch": 4.11,
      "learning_rate": 0.00011777003484320558,
      "loss": 0.7694,
      "step": 1180
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.00011707317073170732,
      "loss": 0.7073,
      "step": 1190
    },
    {
      "epoch": 4.18,
      "learning_rate": 0.00011637630662020907,
      "loss": 0.7813,
      "step": 1200
    },
    {
      "epoch": 4.18,
      "eval_accuracy": 0.7349294045426642,
      "eval_loss": 1.0452630519866943,
      "eval_runtime": 295.364,
      "eval_samples_per_second": 27.576,
      "eval_steps_per_second": 3.45,
      "step": 1200
    },
    {
      "epoch": 4.22,
      "learning_rate": 0.00011567944250871081,
      "loss": 0.7916,
      "step": 1210
    },
    {
      "epoch": 4.25,
      "learning_rate": 0.00011498257839721256,
      "loss": 0.8,
      "step": 1220
    },
    {
      "epoch": 4.29,
      "learning_rate": 0.00011428571428571428,
      "loss": 0.7617,
      "step": 1230
    },
    {
      "epoch": 4.32,
      "learning_rate": 0.00011358885017421603,
      "loss": 0.7524,
      "step": 1240
    },
    {
      "epoch": 4.36,
      "learning_rate": 0.00011289198606271778,
      "loss": 0.7477,
      "step": 1250
    },
    {
      "epoch": 4.39,
      "learning_rate": 0.00011219512195121953,
      "loss": 0.7435,
      "step": 1260
    },
    {
      "epoch": 4.43,
      "learning_rate": 0.00011149825783972125,
      "loss": 0.7493,
      "step": 1270
    },
    {
      "epoch": 4.46,
      "learning_rate": 0.000110801393728223,
      "loss": 0.731,
      "step": 1280
    },
    {
      "epoch": 4.49,
      "learning_rate": 0.00011010452961672475,
      "loss": 0.7566,
      "step": 1290
    },
    {
      "epoch": 4.53,
      "learning_rate": 0.00010940766550522648,
      "loss": 0.7157,
      "step": 1300
    },
    {
      "epoch": 4.53,
      "eval_accuracy": 0.7325966850828729,
      "eval_loss": 1.0333356857299805,
      "eval_runtime": 292.3822,
      "eval_samples_per_second": 27.857,
      "eval_steps_per_second": 3.485,
      "step": 1300
    },
    {
      "epoch": 4.56,
      "learning_rate": 0.00010871080139372823,
      "loss": 0.7028,
      "step": 1310
    },
    {
      "epoch": 4.6,
      "learning_rate": 0.00010801393728222998,
      "loss": 0.7242,
      "step": 1320
    },
    {
      "epoch": 4.63,
      "learning_rate": 0.00010731707317073172,
      "loss": 0.7527,
      "step": 1330
    },
    {
      "epoch": 4.67,
      "learning_rate": 0.00010662020905923345,
      "loss": 0.7347,
      "step": 1340
    },
    {
      "epoch": 4.7,
      "learning_rate": 0.00010592334494773519,
      "loss": 0.745,
      "step": 1350
    },
    {
      "epoch": 4.74,
      "learning_rate": 0.00010522648083623694,
      "loss": 0.7286,
      "step": 1360
    },
    {
      "epoch": 4.77,
      "learning_rate": 0.00010452961672473869,
      "loss": 0.6988,
      "step": 1370
    },
    {
      "epoch": 4.81,
      "learning_rate": 0.00010383275261324041,
      "loss": 0.7018,
      "step": 1380
    },
    {
      "epoch": 4.84,
      "learning_rate": 0.00010313588850174216,
      "loss": 0.7321,
      "step": 1390
    },
    {
      "epoch": 4.88,
      "learning_rate": 0.0001024390243902439,
      "loss": 0.7231,
      "step": 1400
    },
    {
      "epoch": 4.88,
      "eval_accuracy": 0.7453652547575199,
      "eval_loss": 0.9866864681243896,
      "eval_runtime": 327.7217,
      "eval_samples_per_second": 24.853,
      "eval_steps_per_second": 3.109,
      "step": 1400
    },
    {
      "epoch": 4.91,
      "learning_rate": 0.00010174216027874565,
      "loss": 0.7052,
      "step": 1410
    },
    {
      "epoch": 4.95,
      "learning_rate": 0.00010104529616724739,
      "loss": 0.7247,
      "step": 1420
    },
    {
      "epoch": 4.98,
      "learning_rate": 0.00010034843205574914,
      "loss": 0.7218,
      "step": 1430
    },
    {
      "epoch": 5.02,
      "learning_rate": 9.965156794425087e-05,
      "loss": 0.6719,
      "step": 1440
    },
    {
      "epoch": 5.05,
      "learning_rate": 9.895470383275262e-05,
      "loss": 0.6509,
      "step": 1450
    },
    {
      "epoch": 5.09,
      "learning_rate": 9.825783972125437e-05,
      "loss": 0.6253,
      "step": 1460
    },
    {
      "epoch": 5.12,
      "learning_rate": 9.75609756097561e-05,
      "loss": 0.6017,
      "step": 1470
    },
    {
      "epoch": 5.16,
      "learning_rate": 9.686411149825785e-05,
      "loss": 0.6031,
      "step": 1480
    },
    {
      "epoch": 5.19,
      "learning_rate": 9.616724738675959e-05,
      "loss": 0.6391,
      "step": 1490
    },
    {
      "epoch": 5.23,
      "learning_rate": 9.547038327526133e-05,
      "loss": 0.6205,
      "step": 1500
    },
    {
      "epoch": 5.23,
      "eval_accuracy": 0.7446286065070595,
      "eval_loss": 0.9795861840248108,
      "eval_runtime": 296.8791,
      "eval_samples_per_second": 27.435,
      "eval_steps_per_second": 3.432,
      "step": 1500
    },
    {
      "epoch": 5.26,
      "learning_rate": 9.477351916376307e-05,
      "loss": 0.5884,
      "step": 1510
    },
    {
      "epoch": 5.3,
      "learning_rate": 9.407665505226482e-05,
      "loss": 0.62,
      "step": 1520
    },
    {
      "epoch": 5.33,
      "learning_rate": 9.337979094076655e-05,
      "loss": 0.5948,
      "step": 1530
    },
    {
      "epoch": 5.37,
      "learning_rate": 9.26829268292683e-05,
      "loss": 0.5942,
      "step": 1540
    },
    {
      "epoch": 5.4,
      "learning_rate": 9.198606271777003e-05,
      "loss": 0.6308,
      "step": 1550
    },
    {
      "epoch": 5.44,
      "learning_rate": 9.128919860627178e-05,
      "loss": 0.6113,
      "step": 1560
    },
    {
      "epoch": 5.47,
      "learning_rate": 9.059233449477352e-05,
      "loss": 0.6217,
      "step": 1570
    },
    {
      "epoch": 5.51,
      "learning_rate": 8.989547038327526e-05,
      "loss": 0.6375,
      "step": 1580
    },
    {
      "epoch": 5.54,
      "learning_rate": 8.919860627177701e-05,
      "loss": 0.6068,
      "step": 1590
    },
    {
      "epoch": 5.57,
      "learning_rate": 8.850174216027875e-05,
      "loss": 0.5982,
      "step": 1600
    },
    {
      "epoch": 5.57,
      "eval_accuracy": 0.7524861878453039,
      "eval_loss": 0.9573265910148621,
      "eval_runtime": 291.5725,
      "eval_samples_per_second": 27.935,
      "eval_steps_per_second": 3.495,
      "step": 1600
    },
    {
      "epoch": 5.61,
      "learning_rate": 8.78048780487805e-05,
      "loss": 0.5714,
      "step": 1610
    },
    {
      "epoch": 5.64,
      "learning_rate": 8.710801393728223e-05,
      "loss": 0.61,
      "step": 1620
    },
    {
      "epoch": 5.68,
      "learning_rate": 8.641114982578398e-05,
      "loss": 0.5779,
      "step": 1630
    },
    {
      "epoch": 5.71,
      "learning_rate": 8.571428571428571e-05,
      "loss": 0.5917,
      "step": 1640
    },
    {
      "epoch": 5.75,
      "learning_rate": 8.501742160278746e-05,
      "loss": 0.5988,
      "step": 1650
    },
    {
      "epoch": 5.78,
      "learning_rate": 8.43205574912892e-05,
      "loss": 0.5805,
      "step": 1660
    },
    {
      "epoch": 5.82,
      "learning_rate": 8.362369337979094e-05,
      "loss": 0.63,
      "step": 1670
    },
    {
      "epoch": 5.85,
      "learning_rate": 8.292682926829268e-05,
      "loss": 0.572,
      "step": 1680
    },
    {
      "epoch": 5.89,
      "learning_rate": 8.222996515679443e-05,
      "loss": 0.5867,
      "step": 1690
    },
    {
      "epoch": 5.92,
      "learning_rate": 8.153310104529616e-05,
      "loss": 0.6281,
      "step": 1700
    },
    {
      "epoch": 5.92,
      "eval_accuracy": 0.7565377532228361,
      "eval_loss": 0.9512879252433777,
      "eval_runtime": 291.1378,
      "eval_samples_per_second": 27.976,
      "eval_steps_per_second": 3.5,
      "step": 1700
    },
    {
      "epoch": 5.96,
      "learning_rate": 8.083623693379791e-05,
      "loss": 0.5597,
      "step": 1710
    },
    {
      "epoch": 5.99,
      "learning_rate": 8.013937282229966e-05,
      "loss": 0.6012,
      "step": 1720
    },
    {
      "epoch": 6.03,
      "learning_rate": 7.94425087108014e-05,
      "loss": 0.5376,
      "step": 1730
    },
    {
      "epoch": 6.06,
      "learning_rate": 7.874564459930314e-05,
      "loss": 0.4912,
      "step": 1740
    },
    {
      "epoch": 6.1,
      "learning_rate": 7.804878048780489e-05,
      "loss": 0.5127,
      "step": 1750
    },
    {
      "epoch": 6.13,
      "learning_rate": 7.735191637630662e-05,
      "loss": 0.4802,
      "step": 1760
    },
    {
      "epoch": 6.17,
      "learning_rate": 7.665505226480837e-05,
      "loss": 0.5208,
      "step": 1770
    },
    {
      "epoch": 6.2,
      "learning_rate": 7.59581881533101e-05,
      "loss": 0.4987,
      "step": 1780
    },
    {
      "epoch": 6.24,
      "learning_rate": 7.526132404181185e-05,
      "loss": 0.5009,
      "step": 1790
    },
    {
      "epoch": 6.27,
      "learning_rate": 7.456445993031359e-05,
      "loss": 0.5127,
      "step": 1800
    },
    {
      "epoch": 6.27,
      "eval_accuracy": 0.7564149785144261,
      "eval_loss": 0.9356275796890259,
      "eval_runtime": 291.6616,
      "eval_samples_per_second": 27.926,
      "eval_steps_per_second": 3.494,
      "step": 1800
    },
    {
      "epoch": 6.31,
      "learning_rate": 7.386759581881534e-05,
      "loss": 0.4736,
      "step": 1810
    },
    {
      "epoch": 6.34,
      "learning_rate": 7.317073170731707e-05,
      "loss": 0.4973,
      "step": 1820
    },
    {
      "epoch": 6.38,
      "learning_rate": 7.247386759581882e-05,
      "loss": 0.477,
      "step": 1830
    },
    {
      "epoch": 6.41,
      "learning_rate": 7.177700348432055e-05,
      "loss": 0.5001,
      "step": 1840
    },
    {
      "epoch": 6.45,
      "learning_rate": 7.10801393728223e-05,
      "loss": 0.4986,
      "step": 1850
    },
    {
      "epoch": 6.48,
      "learning_rate": 7.038327526132405e-05,
      "loss": 0.5123,
      "step": 1860
    },
    {
      "epoch": 6.52,
      "learning_rate": 6.96864111498258e-05,
      "loss": 0.4856,
      "step": 1870
    },
    {
      "epoch": 6.55,
      "learning_rate": 6.898954703832753e-05,
      "loss": 0.4959,
      "step": 1880
    },
    {
      "epoch": 6.59,
      "learning_rate": 6.829268292682928e-05,
      "loss": 0.4967,
      "step": 1890
    },
    {
      "epoch": 6.62,
      "learning_rate": 6.759581881533101e-05,
      "loss": 0.4718,
      "step": 1900
    },
    {
      "epoch": 6.62,
      "eval_accuracy": 0.7625537139349294,
      "eval_loss": 0.9554521441459656,
      "eval_runtime": 292.36,
      "eval_samples_per_second": 27.859,
      "eval_steps_per_second": 3.485,
      "step": 1900
    },
    {
      "epoch": 6.66,
      "learning_rate": 6.689895470383276e-05,
      "loss": 0.5155,
      "step": 1910
    },
    {
      "epoch": 6.69,
      "learning_rate": 6.62020905923345e-05,
      "loss": 0.4721,
      "step": 1920
    },
    {
      "epoch": 6.72,
      "learning_rate": 6.550522648083623e-05,
      "loss": 0.4835,
      "step": 1930
    },
    {
      "epoch": 6.76,
      "learning_rate": 6.480836236933798e-05,
      "loss": 0.4771,
      "step": 1940
    },
    {
      "epoch": 6.79,
      "learning_rate": 6.411149825783971e-05,
      "loss": 0.5058,
      "step": 1950
    },
    {
      "epoch": 6.83,
      "learning_rate": 6.341463414634146e-05,
      "loss": 0.5301,
      "step": 1960
    },
    {
      "epoch": 6.86,
      "learning_rate": 6.27177700348432e-05,
      "loss": 0.4671,
      "step": 1970
    },
    {
      "epoch": 6.9,
      "learning_rate": 6.202090592334495e-05,
      "loss": 0.4833,
      "step": 1980
    },
    {
      "epoch": 6.93,
      "learning_rate": 6.13240418118467e-05,
      "loss": 0.503,
      "step": 1990
    },
    {
      "epoch": 6.97,
      "learning_rate": 6.0627177700348435e-05,
      "loss": 0.4762,
      "step": 2000
    },
    {
      "epoch": 6.97,
      "eval_accuracy": 0.7695518723143032,
      "eval_loss": 0.8946472406387329,
      "eval_runtime": 290.4161,
      "eval_samples_per_second": 28.046,
      "eval_steps_per_second": 3.509,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "learning_rate": 5.9930313588850176e-05,
      "loss": 0.495,
      "step": 2010
    },
    {
      "epoch": 7.04,
      "learning_rate": 5.9233449477351924e-05,
      "loss": 0.4181,
      "step": 2020
    },
    {
      "epoch": 7.07,
      "learning_rate": 5.853658536585366e-05,
      "loss": 0.3798,
      "step": 2030
    },
    {
      "epoch": 7.11,
      "learning_rate": 5.783972125435541e-05,
      "loss": 0.4163,
      "step": 2040
    },
    {
      "epoch": 7.14,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.441,
      "step": 2050
    },
    {
      "epoch": 7.18,
      "learning_rate": 5.644599303135889e-05,
      "loss": 0.3888,
      "step": 2060
    },
    {
      "epoch": 7.21,
      "learning_rate": 5.5749128919860624e-05,
      "loss": 0.4256,
      "step": 2070
    },
    {
      "epoch": 7.25,
      "learning_rate": 5.505226480836237e-05,
      "loss": 0.4002,
      "step": 2080
    },
    {
      "epoch": 7.28,
      "learning_rate": 5.4355400696864114e-05,
      "loss": 0.4,
      "step": 2090
    },
    {
      "epoch": 7.32,
      "learning_rate": 5.365853658536586e-05,
      "loss": 0.3814,
      "step": 2100
    },
    {
      "epoch": 7.32,
      "eval_accuracy": 0.7636586863106201,
      "eval_loss": 0.9409201145172119,
      "eval_runtime": 291.524,
      "eval_samples_per_second": 27.939,
      "eval_steps_per_second": 3.495,
      "step": 2100
    },
    {
      "epoch": 7.35,
      "learning_rate": 5.2961672473867597e-05,
      "loss": 0.4116,
      "step": 2110
    },
    {
      "epoch": 7.39,
      "learning_rate": 5.2264808362369345e-05,
      "loss": 0.4111,
      "step": 2120
    },
    {
      "epoch": 7.42,
      "learning_rate": 5.156794425087108e-05,
      "loss": 0.4143,
      "step": 2130
    },
    {
      "epoch": 7.46,
      "learning_rate": 5.087108013937283e-05,
      "loss": 0.3869,
      "step": 2140
    },
    {
      "epoch": 7.49,
      "learning_rate": 5.017421602787457e-05,
      "loss": 0.4337,
      "step": 2150
    },
    {
      "epoch": 7.53,
      "learning_rate": 4.947735191637631e-05,
      "loss": 0.3878,
      "step": 2160
    },
    {
      "epoch": 7.56,
      "learning_rate": 4.885017421602788e-05,
      "loss": 0.4246,
      "step": 2170
    },
    {
      "epoch": 7.6,
      "learning_rate": 4.815331010452962e-05,
      "loss": 0.4499,
      "step": 2180
    },
    {
      "epoch": 7.63,
      "learning_rate": 4.745644599303136e-05,
      "loss": 0.4252,
      "step": 2190
    },
    {
      "epoch": 7.67,
      "learning_rate": 4.675958188153311e-05,
      "loss": 0.397,
      "step": 2200
    },
    {
      "epoch": 7.67,
      "eval_accuracy": 0.7673419275629221,
      "eval_loss": 0.9199063777923584,
      "eval_runtime": 293.6426,
      "eval_samples_per_second": 27.738,
      "eval_steps_per_second": 3.47,
      "step": 2200
    },
    {
      "epoch": 7.7,
      "learning_rate": 4.606271777003485e-05,
      "loss": 0.4187,
      "step": 2210
    },
    {
      "epoch": 7.74,
      "learning_rate": 4.536585365853659e-05,
      "loss": 0.3954,
      "step": 2220
    },
    {
      "epoch": 7.77,
      "learning_rate": 4.466898954703833e-05,
      "loss": 0.3953,
      "step": 2230
    },
    {
      "epoch": 7.8,
      "learning_rate": 4.397212543554007e-05,
      "loss": 0.4162,
      "step": 2240
    },
    {
      "epoch": 7.84,
      "learning_rate": 4.3275261324041814e-05,
      "loss": 0.3989,
      "step": 2250
    },
    {
      "epoch": 7.87,
      "learning_rate": 4.2578397212543556e-05,
      "loss": 0.4468,
      "step": 2260
    },
    {
      "epoch": 7.91,
      "learning_rate": 4.1881533101045304e-05,
      "loss": 0.4146,
      "step": 2270
    },
    {
      "epoch": 7.94,
      "learning_rate": 4.1184668989547045e-05,
      "loss": 0.3822,
      "step": 2280
    },
    {
      "epoch": 7.98,
      "learning_rate": 4.0487804878048786e-05,
      "loss": 0.4059,
      "step": 2290
    },
    {
      "epoch": 8.01,
      "learning_rate": 3.979094076655052e-05,
      "loss": 0.3776,
      "step": 2300
    },
    {
      "epoch": 8.01,
      "eval_accuracy": 0.7669736034376918,
      "eval_loss": 0.9196868538856506,
      "eval_runtime": 292.7787,
      "eval_samples_per_second": 27.82,
      "eval_steps_per_second": 3.48,
      "step": 2300
    },
    {
      "epoch": 8.05,
      "learning_rate": 3.909407665505226e-05,
      "loss": 0.3436,
      "step": 2310
    },
    {
      "epoch": 8.08,
      "learning_rate": 3.8397212543554004e-05,
      "loss": 0.3568,
      "step": 2320
    },
    {
      "epoch": 8.12,
      "learning_rate": 3.770034843205575e-05,
      "loss": 0.3505,
      "step": 2330
    },
    {
      "epoch": 8.15,
      "learning_rate": 3.700348432055749e-05,
      "loss": 0.3652,
      "step": 2340
    },
    {
      "epoch": 8.19,
      "learning_rate": 3.6306620209059234e-05,
      "loss": 0.3567,
      "step": 2350
    },
    {
      "epoch": 8.22,
      "learning_rate": 3.5609756097560976e-05,
      "loss": 0.3902,
      "step": 2360
    },
    {
      "epoch": 8.26,
      "learning_rate": 3.491289198606272e-05,
      "loss": 0.3264,
      "step": 2370
    },
    {
      "epoch": 8.29,
      "learning_rate": 3.421602787456446e-05,
      "loss": 0.3866,
      "step": 2380
    },
    {
      "epoch": 8.33,
      "learning_rate": 3.35191637630662e-05,
      "loss": 0.3316,
      "step": 2390
    },
    {
      "epoch": 8.36,
      "learning_rate": 3.282229965156795e-05,
      "loss": 0.3447,
      "step": 2400
    },
    {
      "epoch": 8.36,
      "eval_accuracy": 0.7695518723143032,
      "eval_loss": 0.9099560976028442,
      "eval_runtime": 293.3785,
      "eval_samples_per_second": 27.763,
      "eval_steps_per_second": 3.473,
      "step": 2400
    },
    {
      "epoch": 8.4,
      "learning_rate": 3.212543554006969e-05,
      "loss": 0.3368,
      "step": 2410
    },
    {
      "epoch": 8.43,
      "learning_rate": 3.142857142857143e-05,
      "loss": 0.3633,
      "step": 2420
    },
    {
      "epoch": 8.47,
      "learning_rate": 3.073170731707317e-05,
      "loss": 0.3758,
      "step": 2430
    },
    {
      "epoch": 8.5,
      "learning_rate": 3.0034843205574913e-05,
      "loss": 0.3261,
      "step": 2440
    },
    {
      "epoch": 8.54,
      "learning_rate": 2.9337979094076655e-05,
      "loss": 0.3272,
      "step": 2450
    },
    {
      "epoch": 8.57,
      "learning_rate": 2.86411149825784e-05,
      "loss": 0.3535,
      "step": 2460
    },
    {
      "epoch": 8.61,
      "learning_rate": 2.794425087108014e-05,
      "loss": 0.3336,
      "step": 2470
    },
    {
      "epoch": 8.64,
      "learning_rate": 2.7247386759581882e-05,
      "loss": 0.3334,
      "step": 2480
    },
    {
      "epoch": 8.68,
      "learning_rate": 2.6550522648083627e-05,
      "loss": 0.336,
      "step": 2490
    },
    {
      "epoch": 8.71,
      "learning_rate": 2.5853658536585368e-05,
      "loss": 0.3473,
      "step": 2500
    },
    {
      "epoch": 8.71,
      "eval_accuracy": 0.7691835481890731,
      "eval_loss": 0.9262986183166504,
      "eval_runtime": 292.9844,
      "eval_samples_per_second": 27.8,
      "eval_steps_per_second": 3.478,
      "step": 2500
    },
    {
      "epoch": 8.75,
      "learning_rate": 2.515679442508711e-05,
      "loss": 0.3335,
      "step": 2510
    },
    {
      "epoch": 8.78,
      "learning_rate": 2.445993031358885e-05,
      "loss": 0.3567,
      "step": 2520
    },
    {
      "epoch": 8.82,
      "learning_rate": 2.3763066202090596e-05,
      "loss": 0.3473,
      "step": 2530
    },
    {
      "epoch": 8.85,
      "learning_rate": 2.3066202090592334e-05,
      "loss": 0.3289,
      "step": 2540
    },
    {
      "epoch": 8.89,
      "learning_rate": 2.2369337979094075e-05,
      "loss": 0.3272,
      "step": 2550
    },
    {
      "epoch": 8.92,
      "learning_rate": 2.167247386759582e-05,
      "loss": 0.3401,
      "step": 2560
    },
    {
      "epoch": 8.95,
      "learning_rate": 2.097560975609756e-05,
      "loss": 0.341,
      "step": 2570
    },
    {
      "epoch": 8.99,
      "learning_rate": 2.0278745644599302e-05,
      "loss": 0.3364,
      "step": 2580
    },
    {
      "epoch": 9.02,
      "learning_rate": 1.9581881533101047e-05,
      "loss": 0.3264,
      "step": 2590
    },
    {
      "epoch": 9.06,
      "learning_rate": 1.888501742160279e-05,
      "loss": 0.3242,
      "step": 2600
    },
    {
      "epoch": 9.06,
      "eval_accuracy": 0.7756906077348066,
      "eval_loss": 0.9101601839065552,
      "eval_runtime": 293.3362,
      "eval_samples_per_second": 27.767,
      "eval_steps_per_second": 3.474,
      "step": 2600
    },
    {
      "epoch": 9.09,
      "learning_rate": 1.818815331010453e-05,
      "loss": 0.3211,
      "step": 2610
    },
    {
      "epoch": 9.13,
      "learning_rate": 1.749128919860627e-05,
      "loss": 0.328,
      "step": 2620
    },
    {
      "epoch": 9.16,
      "learning_rate": 1.6794425087108016e-05,
      "loss": 0.3004,
      "step": 2630
    },
    {
      "epoch": 9.2,
      "learning_rate": 1.6097560975609757e-05,
      "loss": 0.2817,
      "step": 2640
    },
    {
      "epoch": 9.23,
      "learning_rate": 1.54006968641115e-05,
      "loss": 0.2847,
      "step": 2650
    },
    {
      "epoch": 9.27,
      "learning_rate": 1.4703832752613242e-05,
      "loss": 0.2653,
      "step": 2660
    },
    {
      "epoch": 9.3,
      "learning_rate": 1.4006968641114985e-05,
      "loss": 0.3139,
      "step": 2670
    },
    {
      "epoch": 9.34,
      "learning_rate": 1.3310104529616726e-05,
      "loss": 0.3207,
      "step": 2680
    },
    {
      "epoch": 9.37,
      "learning_rate": 1.2613240418118469e-05,
      "loss": 0.3138,
      "step": 2690
    },
    {
      "epoch": 9.41,
      "learning_rate": 1.191637630662021e-05,
      "loss": 0.3132,
      "step": 2700
    },
    {
      "epoch": 9.41,
      "eval_accuracy": 0.7720073664825046,
      "eval_loss": 0.9132338166236877,
      "eval_runtime": 293.796,
      "eval_samples_per_second": 27.723,
      "eval_steps_per_second": 3.468,
      "step": 2700
    },
    {
      "epoch": 9.44,
      "learning_rate": 1.1219512195121952e-05,
      "loss": 0.2844,
      "step": 2710
    },
    {
      "epoch": 9.48,
      "learning_rate": 1.0522648083623695e-05,
      "loss": 0.3257,
      "step": 2720
    },
    {
      "epoch": 9.51,
      "learning_rate": 9.825783972125436e-06,
      "loss": 0.3194,
      "step": 2730
    },
    {
      "epoch": 9.55,
      "learning_rate": 9.128919860627178e-06,
      "loss": 0.3099,
      "step": 2740
    },
    {
      "epoch": 9.58,
      "learning_rate": 8.43205574912892e-06,
      "loss": 0.3149,
      "step": 2750
    },
    {
      "epoch": 9.62,
      "learning_rate": 7.735191637630662e-06,
      "loss": 0.2936,
      "step": 2760
    },
    {
      "epoch": 9.65,
      "learning_rate": 7.038327526132404e-06,
      "loss": 0.3089,
      "step": 2770
    },
    {
      "epoch": 9.69,
      "learning_rate": 6.341463414634147e-06,
      "loss": 0.3239,
      "step": 2780
    },
    {
      "epoch": 9.72,
      "learning_rate": 5.644599303135889e-06,
      "loss": 0.2685,
      "step": 2790
    },
    {
      "epoch": 9.76,
      "learning_rate": 4.947735191637631e-06,
      "loss": 0.3238,
      "step": 2800
    },
    {
      "epoch": 9.76,
      "eval_accuracy": 0.7755678330263965,
      "eval_loss": 0.9060993194580078,
      "eval_runtime": 293.0681,
      "eval_samples_per_second": 27.792,
      "eval_steps_per_second": 3.477,
      "step": 2800
    }
  ],
  "max_steps": 2870,
  "num_train_epochs": 10,
  "total_flos": 5.548869192090562e+19,
  "trial_name": null,
  "trial_params": null
}