{
  "best_metric": 0.0159246027469635,
  "best_model_checkpoint": "./vit-driver-drowsiness-detection/checkpoint-5000",
  "epoch": 2.0,
  "global_step": 5988,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.00019966599866399467,
      "loss": 0.3837,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019933199732798934,
      "loss": 0.2449,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019899799599198398,
      "loss": 0.3868,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001986639946559786,
      "loss": 0.2539,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019832999331997328,
      "loss": 0.2597,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019799599198396794,
      "loss": 0.198,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0001976619906479626,
      "loss": 0.2291,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019732798931195727,
      "loss": 0.1267,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019699398797595193,
      "loss": 0.2629,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019665998663994657,
      "loss": 0.2232,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001963259853039412,
      "loss": 0.2117,
      "step": 110
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019599198396793587,
      "loss": 0.1525,
      "step": 120
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019565798263193054,
      "loss": 0.1344,
      "step": 130
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001953239812959252,
      "loss": 0.3125,
      "step": 140
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019498997995991986,
      "loss": 0.2154,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001946559786239145,
      "loss": 0.1723,
      "step": 160
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019432197728790914,
      "loss": 0.2218,
      "step": 170
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001939879759519038,
      "loss": 0.119,
      "step": 180
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019365397461589847,
      "loss": 0.1463,
      "step": 190
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019331997327989313,
      "loss": 0.135,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001929859719438878,
      "loss": 0.3922,
      "step": 210
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019265197060788246,
      "loss": 0.2955,
      "step": 220
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001923179692718771,
      "loss": 0.2092,
      "step": 230
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019198396793587173,
      "loss": 0.1257,
      "step": 240
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001916499665998664,
      "loss": 0.1079,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019131596526386106,
      "loss": 0.1758,
      "step": 260
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019098196392785573,
      "loss": 0.1625,
      "step": 270
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001906479625918504,
      "loss": 0.0636,
      "step": 280
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019031396125584503,
      "loss": 0.4583,
      "step": 290
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00018997995991983967,
      "loss": 0.1824,
      "step": 300
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00018964595858383433,
      "loss": 0.2685,
      "step": 310
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.000189311957247829,
      "loss": 0.2578,
      "step": 320
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00018897795591182366,
      "loss": 0.2213,
      "step": 330
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00018864395457581832,
      "loss": 0.2289,
      "step": 340
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.000188309953239813,
      "loss": 0.1577,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00018797595190380762,
      "loss": 0.1575,
      "step": 360
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00018764195056780226,
      "loss": 0.166,
      "step": 370
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00018734134936539747,
      "loss": 0.3128,
      "step": 380
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00018700734802939214,
      "loss": 0.2045,
      "step": 390
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00018667334669338677,
      "loss": 0.2462,
      "step": 400
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00018633934535738144,
      "loss": 0.206,
      "step": 410
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001860053440213761,
      "loss": 0.1521,
      "step": 420
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00018567134268537074,
      "loss": 0.1971,
      "step": 430
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001853373413493654,
      "loss": 0.3003,
      "step": 440
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00018500334001336007,
      "loss": 0.229,
      "step": 450
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00018466933867735473,
      "loss": 0.1298,
      "step": 460
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00018433533734134937,
      "loss": 0.2014,
      "step": 470
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00018400133600534403,
      "loss": 0.2407,
      "step": 480
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00018366733466933867,
      "loss": 0.1668,
      "step": 490
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00018333333333333334,
      "loss": 0.1504,
      "step": 500
    },
    {
      "epoch": 0.17,
      "eval_accuracy": 0.9539646749342352,
      "eval_loss": 0.11777446419000626,
      "eval_runtime": 99.1546,
      "eval_samples_per_second": 53.674,
      "eval_steps_per_second": 6.717,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000182999331997328,
      "loss": 0.138,
      "step": 510
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00018266533066132266,
      "loss": 0.145,
      "step": 520
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001823313293253173,
      "loss": 0.1636,
      "step": 530
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00018199732798931197,
      "loss": 0.1605,
      "step": 540
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00018166332665330663,
      "loss": 0.1298,
      "step": 550
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00018132932531730127,
      "loss": 0.2324,
      "step": 560
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00018099532398129593,
      "loss": 0.089,
      "step": 570
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001806613226452906,
      "loss": 0.225,
      "step": 580
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018032732130928526,
      "loss": 0.1268,
      "step": 590
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001799933199732799,
      "loss": 0.1399,
      "step": 600
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00017965931863727456,
      "loss": 0.0914,
      "step": 610
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00017932531730126922,
      "loss": 0.0514,
      "step": 620
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00017899131596526386,
      "loss": 0.103,
      "step": 630
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00017865731462925853,
      "loss": 0.1197,
      "step": 640
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001783233132932532,
      "loss": 0.1621,
      "step": 650
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00017798931195724783,
      "loss": 0.1876,
      "step": 660
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001776553106212425,
      "loss": 0.1235,
      "step": 670
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00017732130928523716,
      "loss": 0.1038,
      "step": 680
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001769873079492318,
      "loss": 0.1219,
      "step": 690
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00017665330661322646,
      "loss": 0.0494,
      "step": 700
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00017631930527722112,
      "loss": 0.1744,
      "step": 710
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00017598530394121579,
      "loss": 0.1493,
      "step": 720
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00017565130260521042,
      "loss": 0.1766,
      "step": 730
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001753173012692051,
      "loss": 0.147,
      "step": 740
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00017498329993319975,
      "loss": 0.0798,
      "step": 750
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001746492985971944,
      "loss": 0.0683,
      "step": 760
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00017431529726118905,
      "loss": 0.1262,
      "step": 770
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00017398129592518372,
      "loss": 0.0723,
      "step": 780
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00017364729458917835,
      "loss": 0.2088,
      "step": 790
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00017331329325317302,
      "loss": 0.1181,
      "step": 800
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00017297929191716768,
      "loss": 0.0909,
      "step": 810
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00017264529058116232,
      "loss": 0.1345,
      "step": 820
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017231128924515698,
      "loss": 0.1434,
      "step": 830
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017197728790915165,
      "loss": 0.0939,
      "step": 840
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001716432865731463,
      "loss": 0.0729,
      "step": 850
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017130928523714095,
      "loss": 0.168,
      "step": 860
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017097528390113561,
      "loss": 0.0954,
      "step": 870
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017064128256513028,
      "loss": 0.1173,
      "step": 880
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00017030728122912492,
      "loss": 0.0646,
      "step": 890
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00016997327989311958,
      "loss": 0.0985,
      "step": 900
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00016963927855711424,
      "loss": 0.1258,
      "step": 910
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00016930527722110888,
      "loss": 0.1105,
      "step": 920
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00016897127588510355,
      "loss": 0.124,
      "step": 930
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001686372745490982,
      "loss": 0.0729,
      "step": 940
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00016830327321309287,
      "loss": 0.1236,
      "step": 950
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001679692718770875,
      "loss": 0.1448,
      "step": 960
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00016763527054108218,
      "loss": 0.1346,
      "step": 970
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00016730126920507684,
      "loss": 0.1057,
      "step": 980
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00016696726786907148,
      "loss": 0.0746,
      "step": 990
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00016663326653306614,
      "loss": 0.0581,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "eval_accuracy": 0.9579105599398722,
      "eval_loss": 0.10223273187875748,
      "eval_runtime": 92.9445,
      "eval_samples_per_second": 57.26,
      "eval_steps_per_second": 7.166,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001662992651970608,
      "loss": 0.1906,
      "step": 1010
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00016596526386105544,
      "loss": 0.2125,
      "step": 1020
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001656312625250501,
      "loss": 0.0473,
      "step": 1030
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00016529726118904477,
      "loss": 0.1267,
      "step": 1040
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001649632598530394,
      "loss": 0.0608,
      "step": 1050
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00016462925851703407,
      "loss": 0.0719,
      "step": 1060
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00016429525718102874,
      "loss": 0.1675,
      "step": 1070
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001639612558450234,
      "loss": 0.0754,
      "step": 1080
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00016362725450901804,
      "loss": 0.2477,
      "step": 1090
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001632932531730127,
      "loss": 0.0926,
      "step": 1100
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00016295925183700737,
      "loss": 0.1479,
      "step": 1110
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.000162625250501002,
      "loss": 0.1104,
      "step": 1120
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00016229124916499667,
      "loss": 0.0796,
      "step": 1130
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00016195724782899133,
      "loss": 0.077,
      "step": 1140
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00016162324649298597,
      "loss": 0.0825,
      "step": 1150
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00016128924515698063,
      "loss": 0.1008,
      "step": 1160
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001609552438209753,
      "loss": 0.1497,
      "step": 1170
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00016062124248496993,
      "loss": 0.09,
      "step": 1180
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001602872411489646,
      "loss": 0.0673,
      "step": 1190
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00015995323981295926,
      "loss": 0.0479,
      "step": 1200
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00015961923847695393,
      "loss": 0.1078,
      "step": 1210
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00015928523714094856,
      "loss": 0.0419,
      "step": 1220
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00015895123580494323,
      "loss": 0.1022,
      "step": 1230
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001586172344689379,
      "loss": 0.2065,
      "step": 1240
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00015828323313293253,
      "loss": 0.0798,
      "step": 1250
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001579492317969272,
      "loss": 0.1431,
      "step": 1260
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00015761523046092186,
      "loss": 0.0805,
      "step": 1270
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00015728122912491652,
      "loss": 0.0787,
      "step": 1280
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00015694722778891116,
      "loss": 0.0815,
      "step": 1290
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00015661322645290582,
      "loss": 0.1468,
      "step": 1300
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00015627922511690046,
      "loss": 0.176,
      "step": 1310
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00015594522378089513,
      "loss": 0.1039,
      "step": 1320
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001556112224448898,
      "loss": 0.1102,
      "step": 1330
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00015527722110888445,
      "loss": 0.1112,
      "step": 1340
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001549432197728791,
      "loss": 0.1194,
      "step": 1350
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00015460921843687376,
      "loss": 0.1178,
      "step": 1360
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00015427521710086842,
      "loss": 0.1042,
      "step": 1370
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00015394121576486306,
      "loss": 0.1224,
      "step": 1380
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00015360721442885772,
      "loss": 0.0957,
      "step": 1390
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00015327321309285239,
      "loss": 0.1615,
      "step": 1400
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00015293921175684705,
      "loss": 0.0149,
      "step": 1410
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001526052104208417,
      "loss": 0.1046,
      "step": 1420
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00015227120908483635,
      "loss": 0.0822,
      "step": 1430
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.000151937207748831,
      "loss": 0.0562,
      "step": 1440
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00015160320641282565,
      "loss": 0.0934,
      "step": 1450
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00015126920507682032,
      "loss": 0.076,
      "step": 1460
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00015093520374081498,
      "loss": 0.1111,
      "step": 1470
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00015060120240480962,
      "loss": 0.0998,
      "step": 1480
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00015026720106880428,
      "loss": 0.0688,
      "step": 1490
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00014993319973279895,
      "loss": 0.0415,
      "step": 1500
    },
    {
      "epoch": 0.5,
      "eval_accuracy": 0.9746335963923337,
      "eval_loss": 0.08767460286617279,
      "eval_runtime": 91.9864,
      "eval_samples_per_second": 57.856,
      "eval_steps_per_second": 7.24,
      "step": 1500
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00014959919839679358,
      "loss": 0.0912,
      "step": 1510
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00014926519706078825,
      "loss": 0.1132,
      "step": 1520
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001489311957247829,
      "loss": 0.0564,
      "step": 1530
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00014859719438877758,
      "loss": 0.0951,
      "step": 1540
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014826319305277221,
      "loss": 0.0784,
      "step": 1550
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014792919171676688,
      "loss": 0.0528,
      "step": 1560
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00014759519038076152,
      "loss": 0.0511,
      "step": 1570
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014726118904475618,
      "loss": 0.0287,
      "step": 1580
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00014692718770875084,
      "loss": 0.0658,
      "step": 1590
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001465931863727455,
      "loss": 0.0044,
      "step": 1600
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00014625918503674017,
      "loss": 0.096,
      "step": 1610
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001459251837007348,
      "loss": 0.2206,
      "step": 1620
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00014559118236472947,
      "loss": 0.0855,
      "step": 1630
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001452571810287241,
      "loss": 0.0923,
      "step": 1640
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00014492317969271877,
      "loss": 0.0825,
      "step": 1650
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00014458917835671344,
      "loss": 0.1425,
      "step": 1660
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001442551770207081,
      "loss": 0.0583,
      "step": 1670
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00014392117568470274,
      "loss": 0.0429,
      "step": 1680
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001435871743486974,
      "loss": 0.1406,
      "step": 1690
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00014325317301269204,
      "loss": 0.08,
      "step": 1700
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001429191716766867,
      "loss": 0.0722,
      "step": 1710
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00014258517034068137,
      "loss": 0.0199,
      "step": 1720
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00014225116900467603,
      "loss": 0.1138,
      "step": 1730
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001419171676686707,
      "loss": 0.0515,
      "step": 1740
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00014158316633266534,
      "loss": 0.0228,
      "step": 1750
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00014124916499666,
      "loss": 0.033,
      "step": 1760
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00014091516366065464,
      "loss": 0.0314,
      "step": 1770
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001405811623246493,
      "loss": 0.0835,
      "step": 1780
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00014024716098864397,
      "loss": 0.1252,
      "step": 1790
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00013991315965263863,
      "loss": 0.083,
      "step": 1800
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00013957915831663327,
      "loss": 0.0801,
      "step": 1810
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00013924515698062793,
      "loss": 0.1318,
      "step": 1820
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00013891115564462257,
      "loss": 0.0906,
      "step": 1830
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00013857715430861723,
      "loss": 0.0871,
      "step": 1840
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001382431529726119,
      "loss": 0.086,
      "step": 1850
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00013790915163660656,
      "loss": 0.1238,
      "step": 1860
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00013757515030060123,
      "loss": 0.0535,
      "step": 1870
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00013724114896459586,
      "loss": 0.0443,
      "step": 1880
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00013690714762859053,
      "loss": 0.057,
      "step": 1890
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00013657314629258516,
      "loss": 0.1003,
      "step": 1900
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00013623914495657983,
      "loss": 0.2274,
      "step": 1910
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001359051436205745,
      "loss": 0.0757,
      "step": 1920
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00013557114228456916,
      "loss": 0.0845,
      "step": 1930
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001352371409485638,
      "loss": 0.032,
      "step": 1940
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00013490313961255846,
      "loss": 0.0221,
      "step": 1950
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001345691382765531,
      "loss": 0.0985,
      "step": 1960
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00013423513694054776,
      "loss": 0.0617,
      "step": 1970
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00013390113560454242,
      "loss": 0.0378,
      "step": 1980
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001335671342685371,
      "loss": 0.0427,
      "step": 1990
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00013323313293253175,
      "loss": 0.0487,
      "step": 2000
    },
    {
      "epoch": 0.67,
      "eval_accuracy": 0.9774520856820744,
      "eval_loss": 0.06503941118717194,
      "eval_runtime": 91.3828,
      "eval_samples_per_second": 58.239,
      "eval_steps_per_second": 7.288,
      "step": 2000
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001328991315965264,
      "loss": 0.0452,
      "step": 2010
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00013256513026052105,
      "loss": 0.1043,
      "step": 2020
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001322311289245157,
      "loss": 0.0808,
      "step": 2030
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00013189712758851036,
      "loss": 0.0956,
      "step": 2040
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00013156312625250502,
      "loss": 0.0898,
      "step": 2050
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00013122912491649968,
      "loss": 0.0311,
      "step": 2060
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00013089512358049435,
      "loss": 0.0923,
      "step": 2070
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00013056112224448899,
      "loss": 0.0668,
      "step": 2080
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00013022712090848362,
      "loss": 0.0622,
      "step": 2090
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001298931195724783,
      "loss": 0.0883,
      "step": 2100
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00012955911823647295,
      "loss": 0.0879,
      "step": 2110
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00012922511690046762,
      "loss": 0.0714,
      "step": 2120
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00012889111556446228,
      "loss": 0.0296,
      "step": 2130
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00012855711422845692,
      "loss": 0.0355,
      "step": 2140
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00012822311289245158,
      "loss": 0.1232,
      "step": 2150
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00012788911155644622,
      "loss": 0.2292,
      "step": 2160
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00012755511022044088,
      "loss": 0.0733,
      "step": 2170
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00012722110888443555,
      "loss": 0.0795,
      "step": 2180
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001268871075484302,
      "loss": 0.0814,
      "step": 2190
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00012655310621242487,
      "loss": 0.0262,
      "step": 2200
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001262191048764195,
      "loss": 0.0051,
      "step": 2210
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00012588510354041415,
      "loss": 0.0487,
      "step": 2220
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001255511022044088,
      "loss": 0.0788,
      "step": 2230
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00012521710086840348,
      "loss": 0.0757,
      "step": 2240
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00012488309953239814,
      "loss": 0.0673,
      "step": 2250
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001245490981963928,
      "loss": 0.0391,
      "step": 2260
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00012421509686038744,
      "loss": 0.0056,
      "step": 2270
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001238810955243821,
      "loss": 0.103,
      "step": 2280
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00012354709418837674,
      "loss": 0.0966,
      "step": 2290
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001232130928523714,
      "loss": 0.1016,
      "step": 2300
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00012287909151636607,
      "loss": 0.0309,
      "step": 2310
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00012254509018036074,
      "loss": 0.06,
      "step": 2320
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001222110888443554,
      "loss": 0.0225,
      "step": 2330
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00012187708750835003,
      "loss": 0.1005,
      "step": 2340
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00012154308617234469,
      "loss": 0.0529,
      "step": 2350
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00012120908483633935,
      "loss": 0.1019,
      "step": 2360
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.000120875083500334,
      "loss": 0.0434,
      "step": 2370
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00012054108216432867,
      "loss": 0.03,
      "step": 2380
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00012020708082832333,
      "loss": 0.0998,
      "step": 2390
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00011987307949231798,
      "loss": 0.0385,
      "step": 2400
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00011953907815631262,
      "loss": 0.0348,
      "step": 2410
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00011920507682030728,
      "loss": 0.0323,
      "step": 2420
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00011887107548430194,
      "loss": 0.0914,
      "step": 2430
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001185370741482966,
      "loss": 0.0111,
      "step": 2440
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00011820307281229126,
      "loss": 0.0484,
      "step": 2450
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00011786907147628591,
      "loss": 0.1006,
      "step": 2460
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00011753507014028055,
      "loss": 0.0299,
      "step": 2470
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00011720106880427522,
      "loss": 0.0527,
      "step": 2480
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00011686706746826988,
      "loss": 0.0916,
      "step": 2490
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00011653306613226453,
      "loss": 0.0555,
      "step": 2500
    },
    {
      "epoch": 0.84,
      "eval_accuracy": 0.9785794813979707,
      "eval_loss": 0.05371914803981781,
      "eval_runtime": 91.5705,
      "eval_samples_per_second": 58.119,
      "eval_steps_per_second": 7.273,
      "step": 2500
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001161990647962592,
      "loss": 0.0914,
      "step": 2510
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00011586506346025386,
      "loss": 0.0226,
      "step": 2520
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00011553106212424851,
      "loss": 0.0936,
      "step": 2530
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00011519706078824315,
      "loss": 0.0357,
      "step": 2540
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00011486305945223781,
      "loss": 0.0828,
      "step": 2550
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00011452905811623246,
      "loss": 0.0374,
      "step": 2560
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00011419505678022713,
      "loss": 0.0684,
      "step": 2570
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00011386105544422179,
      "loss": 0.029,
      "step": 2580
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00011352705410821644,
      "loss": 0.0189,
      "step": 2590
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00011319305277221108,
      "loss": 0.1021,
      "step": 2600
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00011285905143620574,
      "loss": 0.0618,
      "step": 2610
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00011252505010020041,
      "loss": 0.0395,
      "step": 2620
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00011219104876419506,
      "loss": 0.0781,
      "step": 2630
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00011185704742818972,
      "loss": 0.0522,
      "step": 2640
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00011152304609218439,
      "loss": 0.0214,
      "step": 2650
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00011118904475617904,
      "loss": 0.0787,
      "step": 2660
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00011085504342017367,
      "loss": 0.0559,
      "step": 2670
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00011052104208416834,
      "loss": 0.0573,
      "step": 2680
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00011018704074816299,
      "loss": 0.0408,
      "step": 2690
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00010985303941215765,
      "loss": 0.0321,
      "step": 2700
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00010951903807615232,
      "loss": 0.1155,
      "step": 2710
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00010918503674014697,
      "loss": 0.1345,
      "step": 2720
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00010885103540414163,
      "loss": 0.0332,
      "step": 2730
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00010851703406813627,
      "loss": 0.0415,
      "step": 2740
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00010818303273213093,
      "loss": 0.0353,
      "step": 2750
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00010784903139612558,
      "loss": 0.025,
      "step": 2760
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00010751503006012025,
      "loss": 0.0112,
      "step": 2770
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00010718102872411491,
      "loss": 0.0153,
      "step": 2780
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00010684702738810956,
      "loss": 0.0078,
      "step": 2790
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001065130260521042,
      "loss": 0.0282,
      "step": 2800
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00010617902471609887,
      "loss": 0.0171,
      "step": 2810
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00010584502338009352,
      "loss": 0.1128,
      "step": 2820
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00010551102204408818,
      "loss": 0.0293,
      "step": 2830
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00010517702070808284,
      "loss": 0.0241,
      "step": 2840
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001048430193720775,
      "loss": 0.2091,
      "step": 2850
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00010454241816967268,
      "loss": 0.057,
      "step": 2860
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00010420841683366733,
      "loss": 0.0085,
      "step": 2870
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.000103874415497662,
      "loss": 0.0383,
      "step": 2880
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00010354041416165665,
      "loss": 0.0131,
      "step": 2890
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00010320641282565131,
      "loss": 0.1179,
      "step": 2900
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00010287241148964597,
      "loss": 0.0122,
      "step": 2910
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00010253841015364062,
      "loss": 0.0375,
      "step": 2920
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00010220440881763526,
      "loss": 0.0353,
      "step": 2930
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00010187040748162993,
      "loss": 0.065,
      "step": 2940
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00010153640614562459,
      "loss": 0.0229,
      "step": 2950
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00010120240480961924,
      "loss": 0.0804,
      "step": 2960
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00010090180360721444,
      "loss": 0.0522,
      "step": 2970
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001005678022712091,
      "loss": 0.1634,
      "step": 2980
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00010023380093520374,
      "loss": 0.0397,
      "step": 2990
    },
    {
      "epoch": 1.0,
      "learning_rate": 9.98997995991984e-05,
      "loss": 0.0279,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9827132656895904,
      "eval_loss": 0.04723888263106346,
      "eval_runtime": 90.6854,
      "eval_samples_per_second": 58.686,
      "eval_steps_per_second": 7.344,
      "step": 3000
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.956579826319305e-05,
      "loss": 0.0893,
      "step": 3010
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.923179692718772e-05,
      "loss": 0.0357,
      "step": 3020
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.889779559118237e-05,
      "loss": 0.0338,
      "step": 3030
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.856379425517702e-05,
      "loss": 0.0205,
      "step": 3040
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.822979291917168e-05,
      "loss": 0.0118,
      "step": 3050
    },
    {
      "epoch": 1.02,
      "learning_rate": 9.789579158316634e-05,
      "loss": 0.0295,
      "step": 3060
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.756179024716099e-05,
      "loss": 0.0199,
      "step": 3070
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.722778891115565e-05,
      "loss": 0.0594,
      "step": 3080
    },
    {
      "epoch": 1.03,
      "learning_rate": 9.68937875751503e-05,
      "loss": 0.0266,
      "step": 3090
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.655978623914495e-05,
      "loss": 0.0981,
      "step": 3100
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.622578490313962e-05,
      "loss": 0.0157,
      "step": 3110
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.589178356713428e-05,
      "loss": 0.0163,
      "step": 3120
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.555778223112893e-05,
      "loss": 0.0089,
      "step": 3130
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.522378089512358e-05,
      "loss": 0.0024,
      "step": 3140
    },
    {
      "epoch": 1.05,
      "learning_rate": 9.488977955911825e-05,
      "loss": 0.0388,
      "step": 3150
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.45557782231129e-05,
      "loss": 0.0382,
      "step": 3160
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.422177688710755e-05,
      "loss": 0.0753,
      "step": 3170
    },
    {
      "epoch": 1.06,
      "learning_rate": 9.388777555110221e-05,
      "loss": 0.0404,
      "step": 3180
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.355377421509686e-05,
      "loss": 0.0182,
      "step": 3190
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.321977287909151e-05,
      "loss": 0.0232,
      "step": 3200
    },
    {
      "epoch": 1.07,
      "learning_rate": 9.288577154308618e-05,
      "loss": 0.0523,
      "step": 3210
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.255177020708084e-05,
      "loss": 0.0103,
      "step": 3220
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.221776887107549e-05,
      "loss": 0.0514,
      "step": 3230
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.188376753507014e-05,
      "loss": 0.0649,
      "step": 3240
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.154976619906481e-05,
      "loss": 0.0591,
      "step": 3250
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.121576486305946e-05,
      "loss": 0.0691,
      "step": 3260
    },
    {
      "epoch": 1.09,
      "learning_rate": 9.088176352705411e-05,
      "loss": 0.0475,
      "step": 3270
    },
    {
      "epoch": 1.1,
      "learning_rate": 9.054776219104877e-05,
      "loss": 0.0152,
      "step": 3280
    },
    {
      "epoch": 1.1,
      "learning_rate": 9.021376085504342e-05,
      "loss": 0.037,
      "step": 3290
    },
    {
      "epoch": 1.1,
      "learning_rate": 8.987975951903807e-05,
      "loss": 0.0166,
      "step": 3300
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.954575818303274e-05,
      "loss": 0.0334,
      "step": 3310
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.921175684702739e-05,
      "loss": 0.0146,
      "step": 3320
    },
    {
      "epoch": 1.11,
      "learning_rate": 8.887775551102205e-05,
      "loss": 0.0316,
      "step": 3330
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.85437541750167e-05,
      "loss": 0.0099,
      "step": 3340
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.820975283901137e-05,
      "loss": 0.0122,
      "step": 3350
    },
    {
      "epoch": 1.12,
      "learning_rate": 8.787575150300602e-05,
      "loss": 0.0067,
      "step": 3360
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.754175016700067e-05,
      "loss": 0.0359,
      "step": 3370
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.720774883099533e-05,
      "loss": 0.0643,
      "step": 3380
    },
    {
      "epoch": 1.13,
      "learning_rate": 8.687374749498998e-05,
      "loss": 0.0018,
      "step": 3390
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.653974615898464e-05,
      "loss": 0.0275,
      "step": 3400
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.62057448229793e-05,
      "loss": 0.0321,
      "step": 3410
    },
    {
      "epoch": 1.14,
      "learning_rate": 8.587174348697395e-05,
      "loss": 0.0214,
      "step": 3420
    },
    {
      "epoch": 1.15,
      "learning_rate": 8.55377421509686e-05,
      "loss": 0.042,
      "step": 3430
    },
    {
      "epoch": 1.15,
      "learning_rate": 8.520374081496326e-05,
      "loss": 0.0317,
      "step": 3440
    },
    {
      "epoch": 1.15,
      "learning_rate": 8.486973947895792e-05,
      "loss": 0.0277,
      "step": 3450
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.453573814295258e-05,
      "loss": 0.0564,
      "step": 3460
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.420173680694723e-05,
      "loss": 0.0127,
      "step": 3470
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.38677354709419e-05,
      "loss": 0.0286,
      "step": 3480
    },
    {
      "epoch": 1.17,
      "learning_rate": 8.353373413493655e-05,
      "loss": 0.0122,
      "step": 3490
    },
    {
      "epoch": 1.17,
      "learning_rate": 8.31997327989312e-05,
      "loss": 0.0139,
      "step": 3500
    },
    {
      "epoch": 1.17,
      "eval_accuracy": 0.9855317549793311,
      "eval_loss": 0.04519040882587433,
      "eval_runtime": 90.7117,
      "eval_samples_per_second": 58.669,
      "eval_steps_per_second": 7.342,
      "step": 3500
    },
    {
      "epoch": 1.17,
      "learning_rate": 8.286573146292586e-05,
      "loss": 0.0098,
      "step": 3510
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.253173012692051e-05,
      "loss": 0.0676,
      "step": 3520
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.219772879091516e-05,
      "loss": 0.0035,
      "step": 3530
    },
    {
      "epoch": 1.18,
      "learning_rate": 8.186372745490983e-05,
      "loss": 0.0323,
      "step": 3540
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.152972611890448e-05,
      "loss": 0.0417,
      "step": 3550
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.119572478289914e-05,
      "loss": 0.0495,
      "step": 3560
    },
    {
      "epoch": 1.19,
      "learning_rate": 8.086172344689379e-05,
      "loss": 0.0199,
      "step": 3570
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.052772211088844e-05,
      "loss": 0.0131,
      "step": 3580
    },
    {
      "epoch": 1.2,
      "learning_rate": 8.01937207748831e-05,
      "loss": 0.0413,
      "step": 3590
    },
    {
      "epoch": 1.2,
      "learning_rate": 7.985971943887776e-05,
      "loss": 0.0587,
      "step": 3600
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.952571810287242e-05,
      "loss": 0.0899,
      "step": 3610
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.919171676686707e-05,
      "loss": 0.0028,
      "step": 3620
    },
    {
      "epoch": 1.21,
      "learning_rate": 7.885771543086172e-05,
      "loss": 0.0159,
      "step": 3630
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.852371409485639e-05,
      "loss": 0.0606,
      "step": 3640
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.818971275885104e-05,
      "loss": 0.0576,
      "step": 3650
    },
    {
      "epoch": 1.22,
      "learning_rate": 7.785571142284569e-05,
      "loss": 0.0505,
      "step": 3660
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.752171008684035e-05,
      "loss": 0.0237,
      "step": 3670
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.7187708750835e-05,
      "loss": 0.0315,
      "step": 3680
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.685370741482967e-05,
      "loss": 0.0512,
      "step": 3690
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.651970607882432e-05,
      "loss": 0.0777,
      "step": 3700
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.618570474281897e-05,
      "loss": 0.0326,
      "step": 3710
    },
    {
      "epoch": 1.24,
      "learning_rate": 7.585170340681363e-05,
      "loss": 0.0206,
      "step": 3720
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.551770207080828e-05,
      "loss": 0.0643,
      "step": 3730
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.518370073480295e-05,
      "loss": 0.0302,
      "step": 3740
    },
    {
      "epoch": 1.25,
      "learning_rate": 7.48496993987976e-05,
      "loss": 0.009,
      "step": 3750
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.451569806279225e-05,
      "loss": 0.0487,
      "step": 3760
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.418169672678691e-05,
      "loss": 0.0771,
      "step": 3770
    },
    {
      "epoch": 1.26,
      "learning_rate": 7.384769539078156e-05,
      "loss": 0.0378,
      "step": 3780
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.351369405477623e-05,
      "loss": 0.0091,
      "step": 3790
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.317969271877088e-05,
      "loss": 0.0059,
      "step": 3800
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.284569138276553e-05,
      "loss": 0.0501,
      "step": 3810
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.25116900467602e-05,
      "loss": 0.0169,
      "step": 3820
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.217768871075485e-05,
      "loss": 0.0208,
      "step": 3830
    },
    {
      "epoch": 1.28,
      "learning_rate": 7.18436873747495e-05,
      "loss": 0.011,
      "step": 3840
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.150968603874416e-05,
      "loss": 0.0308,
      "step": 3850
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.117568470273881e-05,
      "loss": 0.1367,
      "step": 3860
    },
    {
      "epoch": 1.29,
      "learning_rate": 7.084168336673348e-05,
      "loss": 0.0984,
      "step": 3870
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.050768203072813e-05,
      "loss": 0.0335,
      "step": 3880
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.017368069472279e-05,
      "loss": 0.0039,
      "step": 3890
    },
    {
      "epoch": 1.3,
      "learning_rate": 6.983967935871744e-05,
      "loss": 0.0231,
      "step": 3900
    },
    {
      "epoch": 1.31,
      "learning_rate": 6.950567802271209e-05,
      "loss": 0.0609,
      "step": 3910
    },
    {
      "epoch": 1.31,
      "learning_rate": 6.917167668670676e-05,
      "loss": 0.0138,
      "step": 3920
    },
    {
      "epoch": 1.31,
      "learning_rate": 6.88376753507014e-05,
      "loss": 0.061,
      "step": 3930
    },
    {
      "epoch": 1.32,
      "learning_rate": 6.850367401469606e-05,
      "loss": 0.0576,
      "step": 3940
    },
    {
      "epoch": 1.32,
      "learning_rate": 6.816967267869072e-05,
      "loss": 0.0285,
      "step": 3950
    },
    {
      "epoch": 1.32,
      "learning_rate": 6.783567134268537e-05,
      "loss": 0.0169,
      "step": 3960
    },
    {
      "epoch": 1.33,
      "learning_rate": 6.750167000668002e-05,
      "loss": 0.0574,
      "step": 3970
    },
    {
      "epoch": 1.33,
      "learning_rate": 6.716766867067469e-05,
      "loss": 0.1733,
      "step": 3980
    },
    {
      "epoch": 1.33,
      "learning_rate": 6.683366733466934e-05,
      "loss": 0.0473,
      "step": 3990
    },
    {
      "epoch": 1.34,
      "learning_rate": 6.6499665998664e-05,
      "loss": 0.0282,
      "step": 4000
    },
    {
      "epoch": 1.34,
      "eval_accuracy": 0.9877865464111236,
      "eval_loss": 0.035794276744127274,
      "eval_runtime": 91.043,
      "eval_samples_per_second": 58.456,
      "eval_steps_per_second": 7.315,
      "step": 4000
    },
    {
      "epoch": 1.34,
      "learning_rate": 6.616566466265865e-05,
      "loss": 0.0236,
      "step": 4010
    },
    {
      "epoch": 1.34,
      "learning_rate": 6.583166332665332e-05,
      "loss": 0.0372,
      "step": 4020
    },
    {
      "epoch": 1.35,
      "learning_rate": 6.549766199064797e-05,
      "loss": 0.0217,
      "step": 4030
    },
    {
      "epoch": 1.35,
      "learning_rate": 6.516366065464262e-05,
      "loss": 0.053,
      "step": 4040
    },
    {
      "epoch": 1.35,
      "learning_rate": 6.482965931863728e-05,
      "loss": 0.0367,
      "step": 4050
    },
    {
      "epoch": 1.36,
      "learning_rate": 6.449565798263193e-05,
      "loss": 0.0545,
      "step": 4060
    },
    {
      "epoch": 1.36,
      "learning_rate": 6.416165664662658e-05,
      "loss": 0.017,
      "step": 4070
    },
    {
      "epoch": 1.36,
      "learning_rate": 6.382765531062125e-05,
      "loss": 0.0484,
      "step": 4080
    },
    {
      "epoch": 1.37,
      "learning_rate": 6.34936539746159e-05,
      "loss": 0.046,
      "step": 4090
    },
    {
      "epoch": 1.37,
      "learning_rate": 6.315965263861055e-05,
      "loss": 0.0153,
      "step": 4100
    },
    {
      "epoch": 1.37,
      "learning_rate": 6.282565130260521e-05,
      "loss": 0.0107,
      "step": 4110
    },
    {
      "epoch": 1.38,
      "learning_rate": 6.249164996659988e-05,
      "loss": 0.0143,
      "step": 4120
    },
    {
      "epoch": 1.38,
      "learning_rate": 6.215764863059453e-05,
      "loss": 0.0017,
      "step": 4130
    },
    {
      "epoch": 1.38,
      "learning_rate": 6.182364729458918e-05,
      "loss": 0.03,
      "step": 4140
    },
    {
      "epoch": 1.39,
      "learning_rate": 6.148964595858384e-05,
      "loss": 0.015,
      "step": 4150
    },
    {
      "epoch": 1.39,
      "learning_rate": 6.11556446225785e-05,
      "loss": 0.0267,
      "step": 4160
    },
    {
      "epoch": 1.39,
      "learning_rate": 6.082164328657315e-05,
      "loss": 0.0068,
      "step": 4170
    },
    {
      "epoch": 1.4,
      "learning_rate": 6.048764195056781e-05,
      "loss": 0.0085,
      "step": 4180
    },
    {
      "epoch": 1.4,
      "learning_rate": 6.015364061456245e-05,
      "loss": 0.0253,
      "step": 4190
    },
    {
      "epoch": 1.4,
      "learning_rate": 5.981963927855712e-05,
      "loss": 0.076,
      "step": 4200
    },
    {
      "epoch": 1.41,
      "learning_rate": 5.9485637942551775e-05,
      "loss": 0.0656,
      "step": 4210
    },
    {
      "epoch": 1.41,
      "learning_rate": 5.915163660654643e-05,
      "loss": 0.0266,
      "step": 4220
    },
    {
      "epoch": 1.41,
      "learning_rate": 5.881763527054108e-05,
      "loss": 0.0446,
      "step": 4230
    },
    {
      "epoch": 1.42,
      "learning_rate": 5.848363393453574e-05,
      "loss": 0.0554,
      "step": 4240
    },
    {
      "epoch": 1.42,
      "learning_rate": 5.81496325985304e-05,
      "loss": 0.0333,
      "step": 4250
    },
    {
      "epoch": 1.42,
      "learning_rate": 5.781563126252505e-05,
      "loss": 0.0455,
      "step": 4260
    },
    {
      "epoch": 1.43,
      "learning_rate": 5.7481629926519706e-05,
      "loss": 0.0197,
      "step": 4270
    },
    {
      "epoch": 1.43,
      "learning_rate": 5.714762859051437e-05,
      "loss": 0.0158,
      "step": 4280
    },
    {
      "epoch": 1.43,
      "learning_rate": 5.6813627254509014e-05,
      "loss": 0.0106,
      "step": 4290
    },
    {
      "epoch": 1.44,
      "learning_rate": 5.647962591850368e-05,
      "loss": 0.0052,
      "step": 4300
    },
    {
      "epoch": 1.44,
      "learning_rate": 5.6145624582498336e-05,
      "loss": 0.0123,
      "step": 4310
    },
    {
      "epoch": 1.44,
      "learning_rate": 5.581162324649298e-05,
      "loss": 0.0317,
      "step": 4320
    },
    {
      "epoch": 1.45,
      "learning_rate": 5.5477621910487644e-05,
      "loss": 0.0152,
      "step": 4330
    },
    {
      "epoch": 1.45,
      "learning_rate": 5.51436205744823e-05,
      "loss": 0.0531,
      "step": 4340
    },
    {
      "epoch": 1.45,
      "learning_rate": 5.480961923847696e-05,
      "loss": 0.0061,
      "step": 4350
    },
    {
      "epoch": 1.46,
      "learning_rate": 5.447561790247161e-05,
      "loss": 0.0356,
      "step": 4360
    },
    {
      "epoch": 1.46,
      "learning_rate": 5.414161656646627e-05,
      "loss": 0.0444,
      "step": 4370
    },
    {
      "epoch": 1.46,
      "learning_rate": 5.3807615230460925e-05,
      "loss": 0.0525,
      "step": 4380
    },
    {
      "epoch": 1.47,
      "learning_rate": 5.3473613894455576e-05,
      "loss": 0.008,
      "step": 4390
    },
    {
      "epoch": 1.47,
      "learning_rate": 5.313961255845023e-05,
      "loss": 0.0266,
      "step": 4400
    },
    {
      "epoch": 1.47,
      "learning_rate": 5.28056112224449e-05,
      "loss": 0.0429,
      "step": 4410
    },
    {
      "epoch": 1.48,
      "learning_rate": 5.247160988643954e-05,
      "loss": 0.0316,
      "step": 4420
    },
    {
      "epoch": 1.48,
      "learning_rate": 5.2137608550434205e-05,
      "loss": 0.0272,
      "step": 4430
    },
    {
      "epoch": 1.48,
      "learning_rate": 5.180360721442886e-05,
      "loss": 0.0267,
      "step": 4440
    },
    {
      "epoch": 1.49,
      "learning_rate": 5.146960587842352e-05,
      "loss": 0.0173,
      "step": 4450
    },
    {
      "epoch": 1.49,
      "learning_rate": 5.113560454241817e-05,
      "loss": 0.0352,
      "step": 4460
    },
    {
      "epoch": 1.49,
      "learning_rate": 5.080160320641283e-05,
      "loss": 0.0303,
      "step": 4470
    },
    {
      "epoch": 1.5,
      "learning_rate": 5.0467601870407486e-05,
      "loss": 0.04,
      "step": 4480
    },
    {
      "epoch": 1.5,
      "learning_rate": 5.013360053440214e-05,
      "loss": 0.0052,
      "step": 4490
    },
    {
      "epoch": 1.5,
      "learning_rate": 4.9799599198396794e-05,
      "loss": 0.0077,
      "step": 4500
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.9875986471251409,
      "eval_loss": 0.039710693061351776,
      "eval_runtime": 90.7635,
      "eval_samples_per_second": 58.636,
      "eval_steps_per_second": 7.338,
      "step": 4500
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.946559786239145e-05,
      "loss": 0.0448,
      "step": 4510
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.913159652638611e-05,
      "loss": 0.0072,
      "step": 4520
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.879759519038076e-05,
      "loss": 0.0352,
      "step": 4530
    },
    {
      "epoch": 1.52,
      "learning_rate": 4.8463593854375424e-05,
      "loss": 0.038,
      "step": 4540
    },
    {
      "epoch": 1.52,
      "learning_rate": 4.8129592518370075e-05,
      "loss": 0.0278,
      "step": 4550
    },
    {
      "epoch": 1.52,
      "learning_rate": 4.779559118236473e-05,
      "loss": 0.0278,
      "step": 4560
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.746158984635939e-05,
      "loss": 0.0377,
      "step": 4570
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.712758851035404e-05,
      "loss": 0.0228,
      "step": 4580
    },
    {
      "epoch": 1.53,
      "learning_rate": 4.6793587174348705e-05,
      "loss": 0.0364,
      "step": 4590
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.6459585838343355e-05,
      "loss": 0.01,
      "step": 4600
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.612558450233801e-05,
      "loss": 0.0549,
      "step": 4610
    },
    {
      "epoch": 1.54,
      "learning_rate": 4.579158316633267e-05,
      "loss": 0.0302,
      "step": 4620
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.545758183032732e-05,
      "loss": 0.0849,
      "step": 4630
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.5123580494321985e-05,
      "loss": 0.0192,
      "step": 4640
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.4789579158316636e-05,
      "loss": 0.0191,
      "step": 4650
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.4455577822311287e-05,
      "loss": 0.0111,
      "step": 4660
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.412157648630595e-05,
      "loss": 0.0184,
      "step": 4670
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.37875751503006e-05,
      "loss": 0.0156,
      "step": 4680
    },
    {
      "epoch": 1.57,
      "learning_rate": 4.345357381429526e-05,
      "loss": 0.0113,
      "step": 4690
    },
    {
      "epoch": 1.57,
      "learning_rate": 4.3119572478289916e-05,
      "loss": 0.0227,
      "step": 4700
    },
    {
      "epoch": 1.57,
      "learning_rate": 4.278557114228457e-05,
      "loss": 0.0095,
      "step": 4710
    },
    {
      "epoch": 1.58,
      "learning_rate": 4.245156980627923e-05,
      "loss": 0.0487,
      "step": 4720
    },
    {
      "epoch": 1.58,
      "learning_rate": 4.211756847027388e-05,
      "loss": 0.03,
      "step": 4730
    },
    {
      "epoch": 1.58,
      "learning_rate": 4.178356713426854e-05,
      "loss": 0.0045,
      "step": 4740
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.14495657982632e-05,
      "loss": 0.0288,
      "step": 4750
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.111556446225785e-05,
      "loss": 0.0025,
      "step": 4760
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.078156312625251e-05,
      "loss": 0.0338,
      "step": 4770
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.044756179024716e-05,
      "loss": 0.0348,
      "step": 4780
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.011356045424181e-05,
      "loss": 0.0136,
      "step": 4790
    },
    {
      "epoch": 1.6,
      "learning_rate": 3.977955911823648e-05,
      "loss": 0.0375,
      "step": 4800
    },
    {
      "epoch": 1.61,
      "learning_rate": 3.944555778223113e-05,
      "loss": 0.0046,
      "step": 4810
    },
    {
      "epoch": 1.61,
      "learning_rate": 3.9111556446225786e-05,
      "loss": 0.0085,
      "step": 4820
    },
    {
      "epoch": 1.61,
      "learning_rate": 3.877755511022044e-05,
      "loss": 0.0197,
      "step": 4830
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.8443553774215094e-05,
      "loss": 0.0462,
      "step": 4840
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.810955243820976e-05,
      "loss": 0.0095,
      "step": 4850
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.777555110220441e-05,
      "loss": 0.0181,
      "step": 4860
    },
    {
      "epoch": 1.63,
      "learning_rate": 3.7441549766199066e-05,
      "loss": 0.0349,
      "step": 4870
    },
    {
      "epoch": 1.63,
      "learning_rate": 3.7107548430193724e-05,
      "loss": 0.07,
      "step": 4880
    },
    {
      "epoch": 1.63,
      "learning_rate": 3.6773547094188375e-05,
      "loss": 0.0037,
      "step": 4890
    },
    {
      "epoch": 1.64,
      "learning_rate": 3.643954575818304e-05,
      "loss": 0.0087,
      "step": 4900
    },
    {
      "epoch": 1.64,
      "learning_rate": 3.610554442217769e-05,
      "loss": 0.021,
      "step": 4910
    },
    {
      "epoch": 1.64,
      "learning_rate": 3.577154308617235e-05,
      "loss": 0.037,
      "step": 4920
    },
    {
      "epoch": 1.65,
      "learning_rate": 3.5437541750167004e-05,
      "loss": 0.0032,
      "step": 4930
    },
    {
      "epoch": 1.65,
      "learning_rate": 3.5103540414161655e-05,
      "loss": 0.0306,
      "step": 4940
    },
    {
      "epoch": 1.65,
      "learning_rate": 3.476953907815631e-05,
      "loss": 0.0223,
      "step": 4950
    },
    {
      "epoch": 1.66,
      "learning_rate": 3.443553774215097e-05,
      "loss": 0.0463,
      "step": 4960
    },
    {
      "epoch": 1.66,
      "learning_rate": 3.410153640614563e-05,
      "loss": 0.0942,
      "step": 4970
    },
    {
      "epoch": 1.66,
      "learning_rate": 3.3767535070140285e-05,
      "loss": 0.0132,
      "step": 4980
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.3433533734134936e-05,
      "loss": 0.0059,
      "step": 4990
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.309953239812959e-05,
      "loss": 0.0143,
      "step": 5000
    },
    {
      "epoch": 1.67,
      "eval_accuracy": 0.9930477264186396,
      "eval_loss": 0.0159246027469635,
      "eval_runtime": 91.3647,
      "eval_samples_per_second": 58.25,
      "eval_steps_per_second": 7.289,
      "step": 5000
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.276553106212425e-05,
      "loss": 0.0171,
      "step": 5010
    },
    {
      "epoch": 1.68,
      "learning_rate": 3.24315297261189e-05,
      "loss": 0.0192,
      "step": 5020
    },
    {
      "epoch": 1.68,
      "learning_rate": 3.2097528390113566e-05,
      "loss": 0.0009,
      "step": 5030
    },
    {
      "epoch": 1.68,
      "learning_rate": 3.1763527054108216e-05,
      "loss": 0.0159,
      "step": 5040
    },
    {
      "epoch": 1.69,
      "learning_rate": 3.1429525718102874e-05,
      "loss": 0.0193,
      "step": 5050
    },
    {
      "epoch": 1.69,
      "learning_rate": 3.109552438209753e-05,
      "loss": 0.0221,
      "step": 5060
    },
    {
      "epoch": 1.69,
      "learning_rate": 3.076152304609218e-05,
      "loss": 0.0002,
      "step": 5070
    },
    {
      "epoch": 1.7,
      "learning_rate": 3.0427521710086843e-05,
      "loss": 0.0109,
      "step": 5080
    },
    {
      "epoch": 1.7,
      "learning_rate": 3.0093520374081497e-05,
      "loss": 0.0034,
      "step": 5090
    },
    {
      "epoch": 1.7,
      "learning_rate": 2.9759519038076154e-05,
      "loss": 0.0248,
      "step": 5100
    },
    {
      "epoch": 1.71,
      "learning_rate": 2.942551770207081e-05,
      "loss": 0.0187,
      "step": 5110
    },
    {
      "epoch": 1.71,
      "learning_rate": 2.9091516366065462e-05,
      "loss": 0.0404,
      "step": 5120
    },
    {
      "epoch": 1.71,
      "learning_rate": 2.8757515030060123e-05,
      "loss": 0.0064,
      "step": 5130
    },
    {
      "epoch": 1.72,
      "learning_rate": 2.8423513694054777e-05,
      "loss": 0.0719,
      "step": 5140
    },
    {
      "epoch": 1.72,
      "learning_rate": 2.8089512358049435e-05,
      "loss": 0.0505,
      "step": 5150
    },
    {
      "epoch": 1.72,
      "learning_rate": 2.775551102204409e-05,
      "loss": 0.029,
      "step": 5160
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.7421509686038743e-05,
      "loss": 0.027,
      "step": 5170
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.7087508350033404e-05,
      "loss": 0.0151,
      "step": 5180
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.6753507014028058e-05,
      "loss": 0.0392,
      "step": 5190
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.6419505678022715e-05,
      "loss": 0.0015,
      "step": 5200
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.608550434201737e-05,
      "loss": 0.0301,
      "step": 5210
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.5751503006012024e-05,
      "loss": 0.0446,
      "step": 5220
    },
    {
      "epoch": 1.75,
      "learning_rate": 2.541750167000668e-05,
      "loss": 0.0071,
      "step": 5230
    },
    {
      "epoch": 1.75,
      "learning_rate": 2.5083500334001335e-05,
      "loss": 0.0184,
      "step": 5240
    },
    {
      "epoch": 1.75,
      "learning_rate": 2.4749498997995993e-05,
      "loss": 0.007,
      "step": 5250
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.441549766199065e-05,
      "loss": 0.0082,
      "step": 5260
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.4081496325985308e-05,
      "loss": 0.0536,
      "step": 5270
    },
    {
      "epoch": 1.76,
      "learning_rate": 2.3747494989979958e-05,
      "loss": 0.0246,
      "step": 5280
    },
    {
      "epoch": 1.77,
      "learning_rate": 2.3413493653974616e-05,
      "loss": 0.0149,
      "step": 5290
    },
    {
      "epoch": 1.77,
      "learning_rate": 2.3079492317969273e-05,
      "loss": 0.0031,
      "step": 5300
    },
    {
      "epoch": 1.77,
      "learning_rate": 2.274549098196393e-05,
      "loss": 0.0103,
      "step": 5310
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.2411489645958585e-05,
      "loss": 0.0256,
      "step": 5320
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.207748830995324e-05,
      "loss": 0.0025,
      "step": 5330
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.1743486973947896e-05,
      "loss": 0.081,
      "step": 5340
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.1409485637942554e-05,
      "loss": 0.0099,
      "step": 5350
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.1075484301937208e-05,
      "loss": 0.0055,
      "step": 5360
    },
    {
      "epoch": 1.79,
      "learning_rate": 2.0741482965931865e-05,
      "loss": 0.0038,
      "step": 5370
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.040748162992652e-05,
      "loss": 0.0182,
      "step": 5380
    },
    {
      "epoch": 1.8,
      "learning_rate": 2.0073480293921177e-05,
      "loss": 0.0178,
      "step": 5390
    },
    {
      "epoch": 1.8,
      "learning_rate": 1.9739478957915834e-05,
      "loss": 0.0095,
      "step": 5400
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.940547762191049e-05,
      "loss": 0.0092,
      "step": 5410
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.9071476285905142e-05,
      "loss": 0.004,
      "step": 5420
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.87374749498998e-05,
      "loss": 0.02,
      "step": 5430
    },
    {
      "epoch": 1.82,
      "learning_rate": 1.8403473613894457e-05,
      "loss": 0.0152,
      "step": 5440
    },
    {
      "epoch": 1.82,
      "learning_rate": 1.806947227788911e-05,
      "loss": 0.0165,
      "step": 5450
    },
    {
      "epoch": 1.82,
      "learning_rate": 1.773547094188377e-05,
      "loss": 0.0025,
      "step": 5460
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.7401469605878423e-05,
      "loss": 0.0006,
      "step": 5470
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.706746826987308e-05,
      "loss": 0.026,
      "step": 5480
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.6733466933867738e-05,
      "loss": 0.0133,
      "step": 5490
    },
    {
      "epoch": 1.84,
      "learning_rate": 1.6399465597862392e-05,
      "loss": 0.0439,
      "step": 5500
    },
    {
      "epoch": 1.84,
      "eval_accuracy": 0.9930477264186396,
      "eval_loss": 0.016191570088267326,
      "eval_runtime": 90.543,
      "eval_samples_per_second": 58.779,
      "eval_steps_per_second": 7.356,
      "step": 5500
    },
    {
      "epoch": 1.84,
      "learning_rate": 1.606546426185705e-05,
      "loss": 0.0291,
      "step": 5510
    },
    {
      "epoch": 1.84,
      "learning_rate": 1.5731462925851704e-05,
      "loss": 0.0103,
      "step": 5520
    },
    {
      "epoch": 1.85,
      "learning_rate": 1.539746158984636e-05,
      "loss": 0.0081,
      "step": 5530
    },
    {
      "epoch": 1.85,
      "learning_rate": 1.5063460253841017e-05,
      "loss": 0.026,
      "step": 5540
    },
    {
      "epoch": 1.85,
      "learning_rate": 1.4729458917835673e-05,
      "loss": 0.0263,
      "step": 5550
    },
    {
      "epoch": 1.86,
      "learning_rate": 1.4395457581830327e-05,
      "loss": 0.007,
      "step": 5560
    },
    {
      "epoch": 1.86,
      "learning_rate": 1.4061456245824983e-05,
      "loss": 0.001,
      "step": 5570
    },
    {
      "epoch": 1.86,
      "learning_rate": 1.372745490981964e-05,
      "loss": 0.0127,
      "step": 5580
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.3393453573814296e-05,
      "loss": 0.0031,
      "step": 5590
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.3059452237808953e-05,
      "loss": 0.0119,
      "step": 5600
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.2725450901803607e-05,
      "loss": 0.027,
      "step": 5610
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.2391449565798263e-05,
      "loss": 0.0035,
      "step": 5620
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.205744822979292e-05,
      "loss": 0.0469,
      "step": 5630
    },
    {
      "epoch": 1.88,
      "learning_rate": 1.1723446893787576e-05,
      "loss": 0.0171,
      "step": 5640
    },
    {
      "epoch": 1.89,
      "learning_rate": 1.1389445557782232e-05,
      "loss": 0.0218,
      "step": 5650
    },
    {
      "epoch": 1.89,
      "learning_rate": 1.1055444221776888e-05,
      "loss": 0.0074,
      "step": 5660
    },
    {
      "epoch": 1.89,
      "learning_rate": 1.0721442885771544e-05,
      "loss": 0.0011,
      "step": 5670
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.03874415497662e-05,
      "loss": 0.0018,
      "step": 5680
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.0053440213760855e-05,
      "loss": 0.0354,
      "step": 5690
    },
    {
      "epoch": 1.9,
      "learning_rate": 9.719438877755511e-06,
      "loss": 0.0176,
      "step": 5700
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.385437541750168e-06,
      "loss": 0.0019,
      "step": 5710
    },
    {
      "epoch": 1.91,
      "learning_rate": 9.051436205744823e-06,
      "loss": 0.0021,
      "step": 5720
    },
    {
      "epoch": 1.91,
      "learning_rate": 8.71743486973948e-06,
      "loss": 0.0411,
      "step": 5730
    },
    {
      "epoch": 1.92,
      "learning_rate": 8.383433533734134e-06,
      "loss": 0.0037,
      "step": 5740
    },
    {
      "epoch": 1.92,
      "learning_rate": 8.049432197728792e-06,
      "loss": 0.0052,
      "step": 5750
    },
    {
      "epoch": 1.92,
      "learning_rate": 7.715430861723447e-06,
      "loss": 0.0086,
      "step": 5760
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.381429525718103e-06,
      "loss": 0.0295,
      "step": 5770
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.04742818971276e-06,
      "loss": 0.0255,
      "step": 5780
    },
    {
      "epoch": 1.93,
      "learning_rate": 6.713426853707415e-06,
      "loss": 0.0013,
      "step": 5790
    },
    {
      "epoch": 1.94,
      "learning_rate": 6.379425517702071e-06,
      "loss": 0.0318,
      "step": 5800
    },
    {
      "epoch": 1.94,
      "learning_rate": 6.045424181696727e-06,
      "loss": 0.0062,
      "step": 5810
    },
    {
      "epoch": 1.94,
      "learning_rate": 5.711422845691383e-06,
      "loss": 0.0053,
      "step": 5820
    },
    {
      "epoch": 1.95,
      "learning_rate": 5.3774215096860395e-06,
      "loss": 0.0069,
      "step": 5830
    },
    {
      "epoch": 1.95,
      "learning_rate": 5.043420173680695e-06,
      "loss": 0.0144,
      "step": 5840
    },
    {
      "epoch": 1.95,
      "learning_rate": 4.709418837675351e-06,
      "loss": 0.0052,
      "step": 5850
    },
    {
      "epoch": 1.96,
      "learning_rate": 4.375417501670007e-06,
      "loss": 0.0007,
      "step": 5860
    },
    {
      "epoch": 1.96,
      "learning_rate": 4.041416165664663e-06,
      "loss": 0.0134,
      "step": 5870
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.707414829659319e-06,
      "loss": 0.0274,
      "step": 5880
    },
    {
      "epoch": 1.97,
      "learning_rate": 3.373413493653975e-06,
      "loss": 0.0045,
      "step": 5890
    },
    {
      "epoch": 1.97,
      "learning_rate": 3.0394121576486307e-06,
      "loss": 0.0419,
      "step": 5900
    },
    {
      "epoch": 1.97,
      "learning_rate": 2.7054108216432865e-06,
      "loss": 0.0138,
      "step": 5910
    },
    {
      "epoch": 1.98,
      "learning_rate": 2.3714094856379427e-06,
      "loss": 0.1002,
      "step": 5920
    },
    {
      "epoch": 1.98,
      "learning_rate": 2.0374081496325985e-06,
      "loss": 0.0036,
      "step": 5930
    },
    {
      "epoch": 1.98,
      "learning_rate": 1.7034068136272545e-06,
      "loss": 0.0005,
      "step": 5940
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.3694054776219107e-06,
      "loss": 0.0271,
      "step": 5950
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.0354041416165665e-06,
      "loss": 0.0061,
      "step": 5960
    },
    {
      "epoch": 1.99,
      "learning_rate": 7.014028056112224e-07,
      "loss": 0.0123,
      "step": 5970
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.6740146960587843e-07,
      "loss": 0.0132,
      "step": 5980
    },
    {
      "epoch": 2.0,
      "step": 5988,
      "total_flos": 7.422802701202538e+18,
      "train_loss": 0.06747758854398318,
      "train_runtime": 4227.9196,
      "train_samples_per_second": 22.656,
      "train_steps_per_second": 1.416
    }
  ],
  "max_steps": 5988,
  "num_train_epochs": 2,
  "total_flos": 7.422802701202538e+18,
  "trial_name": null,
  "trial_params": null
}