{
  "best_metric": 3.1427292823791504,
  "best_model_checkpoint": "DrishtiSharma/finetuned-ViT-human-action-recognition-v1/checkpoint-100",
  "epoch": 4.0,
  "global_step": 3152,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.00019936548223350255,
      "loss": 2.6364,
      "step": 10
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019873096446700508,
      "loss": 2.4582,
      "step": 20
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019809644670050762,
      "loss": 2.3581,
      "step": 30
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019746192893401016,
      "loss": 2.1244,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019682741116751272,
      "loss": 1.9543,
      "step": 50
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019619289340101523,
      "loss": 1.8044,
      "step": 60
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019555837563451777,
      "loss": 1.6504,
      "step": 70
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019492385786802033,
      "loss": 1.5459,
      "step": 80
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019428934010152284,
      "loss": 1.3911,
      "step": 90
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019365482233502538,
      "loss": 1.4986,
      "step": 100
    },
    {
      "epoch": 0.13,
      "eval_accuracy": 0.07907407407407407,
      "eval_loss": 3.1427292823791504,
      "eval_runtime": 41.775,
      "eval_samples_per_second": 129.264,
      "eval_steps_per_second": 16.158,
      "step": 100
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019302030456852794,
      "loss": 1.3489,
      "step": 110
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019238578680203048,
      "loss": 1.5117,
      "step": 120
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.000191751269035533,
      "loss": 1.3084,
      "step": 130
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019111675126903555,
      "loss": 1.3546,
      "step": 140
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019054568527918783,
      "loss": 1.3806,
      "step": 150
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018991116751269037,
      "loss": 1.1254,
      "step": 160
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001892766497461929,
      "loss": 1.2181,
      "step": 170
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018864213197969544,
      "loss": 1.2078,
      "step": 180
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00018800761421319798,
      "loss": 1.313,
      "step": 190
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00018737309644670052,
      "loss": 1.1929,
      "step": 200
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.0725925925925926,
      "eval_loss": 3.40832257270813,
      "eval_runtime": 42.5099,
      "eval_samples_per_second": 127.029,
      "eval_steps_per_second": 15.879,
      "step": 200
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00018673857868020305,
      "loss": 1.3367,
      "step": 210
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001861040609137056,
      "loss": 1.2115,
      "step": 220
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018546954314720813,
      "loss": 1.3093,
      "step": 230
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018483502538071066,
      "loss": 1.3269,
      "step": 240
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001842005076142132,
      "loss": 1.249,
      "step": 250
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018356598984771574,
      "loss": 1.2936,
      "step": 260
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018293147208121827,
      "loss": 1.1362,
      "step": 270
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001822969543147208,
      "loss": 1.3217,
      "step": 280
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018166243654822338,
      "loss": 1.2232,
      "step": 290
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018102791878172588,
      "loss": 1.2673,
      "step": 300
    },
    {
      "epoch": 0.38,
      "eval_accuracy": 0.07685185185185185,
      "eval_loss": 3.4614734649658203,
      "eval_runtime": 42.2463,
      "eval_samples_per_second": 127.822,
      "eval_steps_per_second": 15.978,
      "step": 300
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018039340101522845,
      "loss": 1.2816,
      "step": 310
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00017975888324873099,
      "loss": 1.2444,
      "step": 320
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001791243654822335,
      "loss": 1.193,
      "step": 330
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00017848984771573606,
      "loss": 1.2228,
      "step": 340
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001778553299492386,
      "loss": 0.9698,
      "step": 350
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017722081218274113,
      "loss": 1.0799,
      "step": 360
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017658629441624367,
      "loss": 1.1013,
      "step": 370
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001759517766497462,
      "loss": 0.9917,
      "step": 380
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017531725888324874,
      "loss": 0.9477,
      "step": 390
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017468274111675128,
      "loss": 0.9805,
      "step": 400
    },
    {
      "epoch": 0.51,
      "eval_accuracy": 0.0824074074074074,
      "eval_loss": 3.9192473888397217,
      "eval_runtime": 43.0917,
      "eval_samples_per_second": 125.314,
      "eval_steps_per_second": 15.664,
      "step": 400
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017404822335025382,
      "loss": 1.0459,
      "step": 410
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017341370558375635,
      "loss": 1.1646,
      "step": 420
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017284263959390863,
      "loss": 1.1712,
      "step": 430
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00017220812182741117,
      "loss": 1.0807,
      "step": 440
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001715736040609137,
      "loss": 1.0306,
      "step": 450
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017093908629441627,
      "loss": 1.2633,
      "step": 460
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00017030456852791878,
      "loss": 1.1891,
      "step": 470
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016967005076142132,
      "loss": 1.0217,
      "step": 480
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016903553299492388,
      "loss": 0.9724,
      "step": 490
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001684010152284264,
      "loss": 1.158,
      "step": 500
    },
    {
      "epoch": 0.63,
      "eval_accuracy": 0.06981481481481482,
      "eval_loss": 4.264822006225586,
      "eval_runtime": 42.7352,
      "eval_samples_per_second": 126.36,
      "eval_steps_per_second": 15.795,
      "step": 500
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016776649746192893,
      "loss": 0.9447,
      "step": 510
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001671319796954315,
      "loss": 1.1464,
      "step": 520
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00016649746192893403,
      "loss": 0.8743,
      "step": 530
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016586294416243654,
      "loss": 1.0659,
      "step": 540
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001652284263959391,
      "loss": 1.0662,
      "step": 550
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016459390862944164,
      "loss": 1.0921,
      "step": 560
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016395939086294415,
      "loss": 0.9721,
      "step": 570
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00016332487309644671,
      "loss": 1.1353,
      "step": 580
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016269035532994925,
      "loss": 1.0537,
      "step": 590
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016205583756345176,
      "loss": 1.2544,
      "step": 600
    },
    {
      "epoch": 0.76,
      "eval_accuracy": 0.05740740740740741,
      "eval_loss": 4.553570747375488,
      "eval_runtime": 42.3697,
      "eval_samples_per_second": 127.45,
      "eval_steps_per_second": 15.931,
      "step": 600
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016142131979695432,
      "loss": 1.0304,
      "step": 610
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016078680203045686,
      "loss": 0.9029,
      "step": 620
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001601522842639594,
      "loss": 0.8909,
      "step": 630
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015951776649746193,
      "loss": 1.0452,
      "step": 640
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015888324873096447,
      "loss": 1.1176,
      "step": 650
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.000158248730964467,
      "loss": 1.036,
      "step": 660
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015761421319796955,
      "loss": 0.9189,
      "step": 670
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015697969543147208,
      "loss": 1.116,
      "step": 680
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00015634517766497462,
      "loss": 1.0224,
      "step": 690
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015571065989847718,
      "loss": 1.0073,
      "step": 700
    },
    {
      "epoch": 0.89,
      "eval_accuracy": 0.08185185185185186,
      "eval_loss": 4.0310468673706055,
      "eval_runtime": 42.4402,
      "eval_samples_per_second": 127.238,
      "eval_steps_per_second": 15.905,
      "step": 700
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001550761421319797,
      "loss": 0.978,
      "step": 710
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015444162436548223,
      "loss": 1.1498,
      "step": 720
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001538071065989848,
      "loss": 1.0484,
      "step": 730
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001531725888324873,
      "loss": 1.0659,
      "step": 740
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015253807106598984,
      "loss": 1.1535,
      "step": 750
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001519035532994924,
      "loss": 0.9928,
      "step": 760
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015126903553299494,
      "loss": 1.063,
      "step": 770
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015063451776649745,
      "loss": 1.0678,
      "step": 780
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.9333,
      "step": 790
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014936548223350255,
      "loss": 0.9315,
      "step": 800
    },
    {
      "epoch": 1.02,
      "eval_accuracy": 0.07018518518518518,
      "eval_loss": 4.515374183654785,
      "eval_runtime": 42.5806,
      "eval_samples_per_second": 126.818,
      "eval_steps_per_second": 15.852,
      "step": 800
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00014873096446700506,
      "loss": 0.8322,
      "step": 810
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00014809644670050762,
      "loss": 1.0184,
      "step": 820
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014746192893401016,
      "loss": 0.8634,
      "step": 830
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001468274111675127,
      "loss": 0.8887,
      "step": 840
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014619289340101523,
      "loss": 0.7927,
      "step": 850
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014555837563451777,
      "loss": 0.9064,
      "step": 860
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001449238578680203,
      "loss": 1.1322,
      "step": 870
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014428934010152285,
      "loss": 0.815,
      "step": 880
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00014365482233502538,
      "loss": 0.806,
      "step": 890
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014302030456852792,
      "loss": 0.9063,
      "step": 900
    },
    {
      "epoch": 1.14,
      "eval_accuracy": 0.06333333333333334,
      "eval_loss": 4.716177940368652,
      "eval_runtime": 42.6095,
      "eval_samples_per_second": 126.732,
      "eval_steps_per_second": 15.842,
      "step": 900
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014238578680203048,
      "loss": 0.7686,
      "step": 910
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.000141751269035533,
      "loss": 0.8536,
      "step": 920
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014111675126903553,
      "loss": 0.9075,
      "step": 930
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0001404822335025381,
      "loss": 0.7441,
      "step": 940
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001398477157360406,
      "loss": 0.7896,
      "step": 950
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013921319796954314,
      "loss": 0.8878,
      "step": 960
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001385786802030457,
      "loss": 0.877,
      "step": 970
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013794416243654824,
      "loss": 1.1784,
      "step": 980
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013730964467005075,
      "loss": 1.011,
      "step": 990
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00013667512690355331,
      "loss": 0.6756,
      "step": 1000
    },
    {
      "epoch": 1.27,
      "eval_accuracy": 0.06259259259259259,
      "eval_loss": 4.648246765136719,
      "eval_runtime": 41.8888,
      "eval_samples_per_second": 128.913,
      "eval_steps_per_second": 16.114,
      "step": 1000
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013604060913705585,
      "loss": 0.7607,
      "step": 1010
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001354060913705584,
      "loss": 1.0117,
      "step": 1020
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013477157360406092,
      "loss": 0.9099,
      "step": 1030
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013413705583756346,
      "loss": 0.8576,
      "step": 1040
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.000133502538071066,
      "loss": 0.9209,
      "step": 1050
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013286802030456854,
      "loss": 0.709,
      "step": 1060
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013223350253807107,
      "loss": 0.9996,
      "step": 1070
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001315989847715736,
      "loss": 0.9422,
      "step": 1080
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013096446700507615,
      "loss": 0.9695,
      "step": 1090
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00013032994923857868,
      "loss": 1.0239,
      "step": 1100
    },
    {
      "epoch": 1.4,
      "eval_accuracy": 0.06351851851851852,
      "eval_loss": 4.643730640411377,
      "eval_runtime": 42.7218,
      "eval_samples_per_second": 126.399,
      "eval_steps_per_second": 15.8,
      "step": 1100
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012969543147208122,
      "loss": 0.7743,
      "step": 1110
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012906091370558376,
      "loss": 0.9336,
      "step": 1120
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001284263959390863,
      "loss": 0.9458,
      "step": 1130
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012779187817258883,
      "loss": 0.6924,
      "step": 1140
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001271573604060914,
      "loss": 0.9117,
      "step": 1150
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001265228426395939,
      "loss": 0.8666,
      "step": 1160
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00012588832487309644,
      "loss": 0.7404,
      "step": 1170
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.000125253807106599,
      "loss": 0.8045,
      "step": 1180
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012461928934010151,
      "loss": 0.8253,
      "step": 1190
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00012398477157360405,
      "loss": 0.7634,
      "step": 1200
    },
    {
      "epoch": 1.52,
      "eval_accuracy": 0.07518518518518519,
      "eval_loss": 4.562452793121338,
      "eval_runtime": 42.3541,
      "eval_samples_per_second": 127.496,
      "eval_steps_per_second": 15.937,
      "step": 1200
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00012335025380710661,
      "loss": 0.8144,
      "step": 1210
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012271573604060915,
      "loss": 0.9033,
      "step": 1220
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001220812182741117,
      "loss": 0.9167,
      "step": 1230
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012144670050761423,
      "loss": 0.6407,
      "step": 1240
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012081218274111676,
      "loss": 0.7116,
      "step": 1250
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00012017766497461929,
      "loss": 0.9067,
      "step": 1260
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00011954314720812184,
      "loss": 0.8048,
      "step": 1270
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011890862944162437,
      "loss": 0.7809,
      "step": 1280
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011827411167512692,
      "loss": 0.7059,
      "step": 1290
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011763959390862945,
      "loss": 0.8365,
      "step": 1300
    },
    {
      "epoch": 1.65,
      "eval_accuracy": 0.05611111111111111,
      "eval_loss": 4.9912309646606445,
      "eval_runtime": 42.3693,
      "eval_samples_per_second": 127.451,
      "eval_steps_per_second": 15.931,
      "step": 1300
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011700507614213198,
      "loss": 0.7949,
      "step": 1310
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011637055837563453,
      "loss": 0.7233,
      "step": 1320
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011573604060913706,
      "loss": 0.7498,
      "step": 1330
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001151015228426396,
      "loss": 0.8727,
      "step": 1340
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011446700507614214,
      "loss": 0.7559,
      "step": 1350
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011383248730964468,
      "loss": 0.9523,
      "step": 1360
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0001131979695431472,
      "loss": 0.8862,
      "step": 1370
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011256345177664975,
      "loss": 0.7775,
      "step": 1380
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011192893401015229,
      "loss": 0.7125,
      "step": 1390
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011129441624365481,
      "loss": 0.8979,
      "step": 1400
    },
    {
      "epoch": 1.78,
      "eval_accuracy": 0.035555555555555556,
      "eval_loss": 5.173853397369385,
      "eval_runtime": 42.3905,
      "eval_samples_per_second": 127.387,
      "eval_steps_per_second": 15.923,
      "step": 1400
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011065989847715736,
      "loss": 0.6235,
      "step": 1410
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011008883248730966,
      "loss": 0.8089,
      "step": 1420
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010945431472081218,
      "loss": 0.8081,
      "step": 1430
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010881979695431472,
      "loss": 0.8359,
      "step": 1440
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010818527918781727,
      "loss": 0.9462,
      "step": 1450
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001075507614213198,
      "loss": 0.7525,
      "step": 1460
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00010691624365482233,
      "loss": 0.9649,
      "step": 1470
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010628172588832488,
      "loss": 0.9125,
      "step": 1480
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010564720812182743,
      "loss": 0.8711,
      "step": 1490
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010501269035532994,
      "loss": 0.9448,
      "step": 1500
    },
    {
      "epoch": 1.9,
      "eval_accuracy": 0.05407407407407407,
      "eval_loss": 4.894594669342041,
      "eval_runtime": 42.8708,
      "eval_samples_per_second": 125.96,
      "eval_steps_per_second": 15.745,
      "step": 1500
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010437817258883249,
      "loss": 0.8529,
      "step": 1510
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010374365482233504,
      "loss": 0.9543,
      "step": 1520
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010310913705583758,
      "loss": 0.6497,
      "step": 1530
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001024746192893401,
      "loss": 0.9036,
      "step": 1540
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010184010152284265,
      "loss": 0.8185,
      "step": 1550
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010120558375634519,
      "loss": 0.8283,
      "step": 1560
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010057106598984771,
      "loss": 0.8138,
      "step": 1570
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.993654822335026e-05,
      "loss": 0.6561,
      "step": 1580
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.93020304568528e-05,
      "loss": 0.6175,
      "step": 1590
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.866751269035533e-05,
      "loss": 0.697,
      "step": 1600
    },
    {
      "epoch": 2.03,
      "eval_accuracy": 0.07407407407407407,
      "eval_loss": 4.951629638671875,
      "eval_runtime": 42.4135,
      "eval_samples_per_second": 127.318,
      "eval_steps_per_second": 15.915,
      "step": 1600
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.803299492385787e-05,
      "loss": 0.6009,
      "step": 1610
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.739847715736041e-05,
      "loss": 0.5976,
      "step": 1620
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.676395939086295e-05,
      "loss": 0.7435,
      "step": 1630
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.61294416243655e-05,
      "loss": 0.4801,
      "step": 1640
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.549492385786802e-05,
      "loss": 0.6946,
      "step": 1650
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.486040609137057e-05,
      "loss": 0.7251,
      "step": 1660
    },
    {
      "epoch": 2.12,
      "learning_rate": 9.42258883248731e-05,
      "loss": 0.646,
      "step": 1670
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.359137055837563e-05,
      "loss": 0.7464,
      "step": 1680
    },
    {
      "epoch": 2.14,
      "learning_rate": 9.295685279187818e-05,
      "loss": 0.7084,
      "step": 1690
    },
    {
      "epoch": 2.16,
      "learning_rate": 9.232233502538072e-05,
      "loss": 0.7861,
      "step": 1700
    },
    {
      "epoch": 2.16,
      "eval_accuracy": 0.07759259259259259,
      "eval_loss": 5.008974075317383,
      "eval_runtime": 42.9257,
      "eval_samples_per_second": 125.799,
      "eval_steps_per_second": 15.725,
      "step": 1700
    },
    {
      "epoch": 2.17,
      "learning_rate": 9.168781725888325e-05,
      "loss": 0.4633,
      "step": 1710
    },
    {
      "epoch": 2.18,
      "learning_rate": 9.105329949238579e-05,
      "loss": 0.6661,
      "step": 1720
    },
    {
      "epoch": 2.2,
      "learning_rate": 9.041878172588833e-05,
      "loss": 0.6019,
      "step": 1730
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.978426395939086e-05,
      "loss": 0.522,
      "step": 1740
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.91497461928934e-05,
      "loss": 0.7008,
      "step": 1750
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.851522842639595e-05,
      "loss": 0.6761,
      "step": 1760
    },
    {
      "epoch": 2.25,
      "learning_rate": 8.788071065989847e-05,
      "loss": 0.844,
      "step": 1770
    },
    {
      "epoch": 2.26,
      "learning_rate": 8.724619289340102e-05,
      "loss": 0.6207,
      "step": 1780
    },
    {
      "epoch": 2.27,
      "learning_rate": 8.661167512690356e-05,
      "loss": 0.613,
      "step": 1790
    },
    {
      "epoch": 2.28,
      "learning_rate": 8.597715736040608e-05,
      "loss": 0.6404,
      "step": 1800
    },
    {
      "epoch": 2.28,
      "eval_accuracy": 0.06425925925925927,
      "eval_loss": 5.390510559082031,
      "eval_runtime": 41.9103,
      "eval_samples_per_second": 128.847,
      "eval_steps_per_second": 16.106,
      "step": 1800
    },
    {
      "epoch": 2.3,
      "learning_rate": 8.534263959390864e-05,
      "loss": 0.7611,
      "step": 1810
    },
    {
      "epoch": 2.31,
      "learning_rate": 8.470812182741117e-05,
      "loss": 0.5113,
      "step": 1820
    },
    {
      "epoch": 2.32,
      "learning_rate": 8.407360406091371e-05,
      "loss": 0.6598,
      "step": 1830
    },
    {
      "epoch": 2.34,
      "learning_rate": 8.343908629441625e-05,
      "loss": 0.6591,
      "step": 1840
    },
    {
      "epoch": 2.35,
      "learning_rate": 8.28045685279188e-05,
      "loss": 0.7027,
      "step": 1850
    },
    {
      "epoch": 2.36,
      "learning_rate": 8.217005076142132e-05,
      "loss": 0.7335,
      "step": 1860
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.153553299492386e-05,
      "loss": 0.5697,
      "step": 1870
    },
    {
      "epoch": 2.39,
      "learning_rate": 8.09010152284264e-05,
      "loss": 0.7282,
      "step": 1880
    },
    {
      "epoch": 2.4,
      "learning_rate": 8.026649746192893e-05,
      "loss": 0.5357,
      "step": 1890
    },
    {
      "epoch": 2.41,
      "learning_rate": 7.963197969543148e-05,
      "loss": 0.7939,
      "step": 1900
    },
    {
      "epoch": 2.41,
      "eval_accuracy": 0.10148148148148148,
      "eval_loss": 4.915863513946533,
      "eval_runtime": 42.4014,
      "eval_samples_per_second": 127.354,
      "eval_steps_per_second": 15.919,
      "step": 1900
    },
    {
      "epoch": 2.42,
      "learning_rate": 7.899746192893402e-05,
      "loss": 0.6813,
      "step": 1910
    },
    {
      "epoch": 2.44,
      "learning_rate": 7.836294416243654e-05,
      "loss": 0.5975,
      "step": 1920
    },
    {
      "epoch": 2.45,
      "learning_rate": 7.772842639593909e-05,
      "loss": 0.5243,
      "step": 1930
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.709390862944163e-05,
      "loss": 0.6133,
      "step": 1940
    },
    {
      "epoch": 2.47,
      "learning_rate": 7.645939086294416e-05,
      "loss": 0.6493,
      "step": 1950
    },
    {
      "epoch": 2.49,
      "learning_rate": 7.58248730964467e-05,
      "loss": 0.6631,
      "step": 1960
    },
    {
      "epoch": 2.5,
      "learning_rate": 7.519035532994925e-05,
      "loss": 0.593,
      "step": 1970
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.455583756345177e-05,
      "loss": 0.6585,
      "step": 1980
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.392131979695431e-05,
      "loss": 0.5895,
      "step": 1990
    },
    {
      "epoch": 2.54,
      "learning_rate": 7.328680203045686e-05,
      "loss": 0.6331,
      "step": 2000
    },
    {
      "epoch": 2.54,
      "eval_accuracy": 0.058888888888888886,
      "eval_loss": 5.3082709312438965,
      "eval_runtime": 42.5362,
      "eval_samples_per_second": 126.951,
      "eval_steps_per_second": 15.869,
      "step": 2000
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.265228426395939e-05,
      "loss": 0.7196,
      "step": 2010
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.201776649746194e-05,
      "loss": 0.6474,
      "step": 2020
    },
    {
      "epoch": 2.58,
      "learning_rate": 7.138324873096447e-05,
      "loss": 0.6758,
      "step": 2030
    },
    {
      "epoch": 2.59,
      "learning_rate": 7.074873096446701e-05,
      "loss": 0.6885,
      "step": 2040
    },
    {
      "epoch": 2.6,
      "learning_rate": 7.011421319796955e-05,
      "loss": 0.6022,
      "step": 2050
    },
    {
      "epoch": 2.61,
      "learning_rate": 6.947969543147208e-05,
      "loss": 0.9147,
      "step": 2060
    },
    {
      "epoch": 2.63,
      "learning_rate": 6.884517766497462e-05,
      "loss": 0.7084,
      "step": 2070
    },
    {
      "epoch": 2.64,
      "learning_rate": 6.821065989847716e-05,
      "loss": 0.5282,
      "step": 2080
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.75761421319797e-05,
      "loss": 0.6432,
      "step": 2090
    },
    {
      "epoch": 2.66,
      "learning_rate": 6.694162436548223e-05,
      "loss": 0.6082,
      "step": 2100
    },
    {
      "epoch": 2.66,
      "eval_accuracy": 0.08574074074074074,
      "eval_loss": 4.853830337524414,
      "eval_runtime": 42.042,
      "eval_samples_per_second": 128.443,
      "eval_steps_per_second": 16.055,
      "step": 2100
    },
    {
      "epoch": 2.68,
      "learning_rate": 6.630710659898477e-05,
      "loss": 0.785,
      "step": 2110
    },
    {
      "epoch": 2.69,
      "learning_rate": 6.567258883248732e-05,
      "loss": 0.5666,
      "step": 2120
    },
    {
      "epoch": 2.7,
      "learning_rate": 6.503807106598985e-05,
      "loss": 0.7502,
      "step": 2130
    },
    {
      "epoch": 2.72,
      "learning_rate": 6.440355329949239e-05,
      "loss": 0.6761,
      "step": 2140
    },
    {
      "epoch": 2.73,
      "learning_rate": 6.376903553299493e-05,
      "loss": 0.7513,
      "step": 2150
    },
    {
      "epoch": 2.74,
      "learning_rate": 6.313451776649746e-05,
      "loss": 0.6386,
      "step": 2160
    },
    {
      "epoch": 2.75,
      "learning_rate": 6.25e-05,
      "loss": 0.6444,
      "step": 2170
    },
    {
      "epoch": 2.77,
      "learning_rate": 6.186548223350254e-05,
      "loss": 0.6356,
      "step": 2180
    },
    {
      "epoch": 2.78,
      "learning_rate": 6.123096446700507e-05,
      "loss": 0.634,
      "step": 2190
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.059644670050761e-05,
      "loss": 0.6229,
      "step": 2200
    },
    {
      "epoch": 2.79,
      "eval_accuracy": 0.06888888888888889,
      "eval_loss": 5.3085527420043945,
      "eval_runtime": 42.2815,
      "eval_samples_per_second": 127.715,
      "eval_steps_per_second": 15.964,
      "step": 2200
    },
    {
      "epoch": 2.8,
      "learning_rate": 5.9961928934010155e-05,
      "loss": 0.7879,
      "step": 2210
    },
    {
      "epoch": 2.82,
      "learning_rate": 5.932741116751269e-05,
      "loss": 0.832,
      "step": 2220
    },
    {
      "epoch": 2.83,
      "learning_rate": 5.8692893401015236e-05,
      "loss": 0.6145,
      "step": 2230
    },
    {
      "epoch": 2.84,
      "learning_rate": 5.805837563451777e-05,
      "loss": 0.5856,
      "step": 2240
    },
    {
      "epoch": 2.86,
      "learning_rate": 5.74238578680203e-05,
      "loss": 0.722,
      "step": 2250
    },
    {
      "epoch": 2.87,
      "learning_rate": 5.6789340101522846e-05,
      "loss": 0.7173,
      "step": 2260
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.615482233502538e-05,
      "loss": 0.6153,
      "step": 2270
    },
    {
      "epoch": 2.89,
      "learning_rate": 5.552030456852793e-05,
      "loss": 0.5231,
      "step": 2280
    },
    {
      "epoch": 2.91,
      "learning_rate": 5.488578680203046e-05,
      "loss": 0.5091,
      "step": 2290
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.4251269035532994e-05,
      "loss": 0.6964,
      "step": 2300
    },
    {
      "epoch": 2.92,
      "eval_accuracy": 0.0712962962962963,
      "eval_loss": 5.374546051025391,
      "eval_runtime": 42.4217,
      "eval_samples_per_second": 127.293,
      "eval_steps_per_second": 15.912,
      "step": 2300
    },
    {
      "epoch": 2.93,
      "learning_rate": 5.361675126903554e-05,
      "loss": 0.5486,
      "step": 2310
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.298223350253807e-05,
      "loss": 0.7241,
      "step": 2320
    },
    {
      "epoch": 2.96,
      "learning_rate": 5.234771573604062e-05,
      "loss": 0.6744,
      "step": 2330
    },
    {
      "epoch": 2.97,
      "learning_rate": 5.171319796954315e-05,
      "loss": 0.7482,
      "step": 2340
    },
    {
      "epoch": 2.98,
      "learning_rate": 5.107868020304569e-05,
      "loss": 0.5178,
      "step": 2350
    },
    {
      "epoch": 2.99,
      "learning_rate": 5.044416243654823e-05,
      "loss": 0.6667,
      "step": 2360
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.9809644670050765e-05,
      "loss": 0.5947,
      "step": 2370
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.91751269035533e-05,
      "loss": 0.4845,
      "step": 2380
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.8540609137055845e-05,
      "loss": 0.5299,
      "step": 2390
    },
    {
      "epoch": 3.05,
      "learning_rate": 4.7906091370558375e-05,
      "loss": 0.5246,
      "step": 2400
    },
    {
      "epoch": 3.05,
      "eval_accuracy": 0.07962962962962963,
      "eval_loss": 5.036931991577148,
      "eval_runtime": 42.4212,
      "eval_samples_per_second": 127.295,
      "eval_steps_per_second": 15.912,
      "step": 2400
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.727157360406091e-05,
      "loss": 0.5824,
      "step": 2410
    },
    {
      "epoch": 3.07,
      "learning_rate": 4.6637055837563456e-05,
      "loss": 0.5149,
      "step": 2420
    },
    {
      "epoch": 3.08,
      "learning_rate": 4.600253807106599e-05,
      "loss": 0.5117,
      "step": 2430
    },
    {
      "epoch": 3.1,
      "learning_rate": 4.536802030456853e-05,
      "loss": 0.4887,
      "step": 2440
    },
    {
      "epoch": 3.11,
      "learning_rate": 4.473350253807107e-05,
      "loss": 0.5999,
      "step": 2450
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.40989847715736e-05,
      "loss": 0.4309,
      "step": 2460
    },
    {
      "epoch": 3.13,
      "learning_rate": 4.346446700507614e-05,
      "loss": 0.5065,
      "step": 2470
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.282994923857868e-05,
      "loss": 0.5616,
      "step": 2480
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.219543147208122e-05,
      "loss": 0.4315,
      "step": 2490
    },
    {
      "epoch": 3.17,
      "learning_rate": 4.156091370558376e-05,
      "loss": 0.6097,
      "step": 2500
    },
    {
      "epoch": 3.17,
      "eval_accuracy": 0.07425925925925926,
      "eval_loss": 5.293472766876221,
      "eval_runtime": 42.2861,
      "eval_samples_per_second": 127.702,
      "eval_steps_per_second": 15.963,
      "step": 2500
    },
    {
      "epoch": 3.19,
      "learning_rate": 4.09263959390863e-05,
      "loss": 0.5502,
      "step": 2510
    },
    {
      "epoch": 3.2,
      "learning_rate": 4.029187817258883e-05,
      "loss": 0.7232,
      "step": 2520
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.965736040609137e-05,
      "loss": 0.5662,
      "step": 2530
    },
    {
      "epoch": 3.22,
      "learning_rate": 3.902284263959391e-05,
      "loss": 0.5821,
      "step": 2540
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.838832487309645e-05,
      "loss": 0.4186,
      "step": 2550
    },
    {
      "epoch": 3.25,
      "learning_rate": 3.7753807106598985e-05,
      "loss": 0.4694,
      "step": 2560
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.711928934010153e-05,
      "loss": 0.5519,
      "step": 2570
    },
    {
      "epoch": 3.27,
      "learning_rate": 3.6484771573604065e-05,
      "loss": 0.5724,
      "step": 2580
    },
    {
      "epoch": 3.29,
      "learning_rate": 3.5850253807106595e-05,
      "loss": 0.6541,
      "step": 2590
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.521573604060914e-05,
      "loss": 0.5778,
      "step": 2600
    },
    {
      "epoch": 3.3,
      "eval_accuracy": 0.07092592592592592,
      "eval_loss": 5.5430707931518555,
      "eval_runtime": 42.7306,
      "eval_samples_per_second": 126.373,
      "eval_steps_per_second": 15.797,
      "step": 2600
    },
    {
      "epoch": 3.31,
      "learning_rate": 3.4581218274111676e-05,
      "loss": 0.6211,
      "step": 2610
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.394670050761421e-05,
      "loss": 0.5961,
      "step": 2620
    },
    {
      "epoch": 3.34,
      "learning_rate": 3.3312182741116756e-05,
      "loss": 0.5075,
      "step": 2630
    },
    {
      "epoch": 3.35,
      "learning_rate": 3.267766497461929e-05,
      "loss": 0.5977,
      "step": 2640
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.204314720812183e-05,
      "loss": 0.4841,
      "step": 2650
    },
    {
      "epoch": 3.38,
      "learning_rate": 3.1408629441624366e-05,
      "loss": 0.4477,
      "step": 2660
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.07741116751269e-05,
      "loss": 0.4101,
      "step": 2670
    },
    {
      "epoch": 3.4,
      "learning_rate": 3.0139593908629443e-05,
      "loss": 0.4422,
      "step": 2680
    },
    {
      "epoch": 3.41,
      "learning_rate": 2.9505076142131984e-05,
      "loss": 0.5842,
      "step": 2690
    },
    {
      "epoch": 3.43,
      "learning_rate": 2.887055837563452e-05,
      "loss": 0.4196,
      "step": 2700
    },
    {
      "epoch": 3.43,
      "eval_accuracy": 0.07592592592592592,
      "eval_loss": 5.55080509185791,
      "eval_runtime": 42.1099,
      "eval_samples_per_second": 128.236,
      "eval_steps_per_second": 16.029,
      "step": 2700
    },
    {
      "epoch": 3.44,
      "learning_rate": 2.823604060913706e-05,
      "loss": 0.6589,
      "step": 2710
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.7601522842639594e-05,
      "loss": 0.4259,
      "step": 2720
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.696700507614213e-05,
      "loss": 0.4385,
      "step": 2730
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.633248730964467e-05,
      "loss": 0.5176,
      "step": 2740
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.569796954314721e-05,
      "loss": 0.6453,
      "step": 2750
    },
    {
      "epoch": 3.5,
      "learning_rate": 2.5063451776649748e-05,
      "loss": 0.5074,
      "step": 2760
    },
    {
      "epoch": 3.52,
      "learning_rate": 2.4428934010152285e-05,
      "loss": 0.518,
      "step": 2770
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.3794416243654825e-05,
      "loss": 0.4455,
      "step": 2780
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.315989847715736e-05,
      "loss": 0.6174,
      "step": 2790
    },
    {
      "epoch": 3.55,
      "learning_rate": 2.25253807106599e-05,
      "loss": 0.5495,
      "step": 2800
    },
    {
      "epoch": 3.55,
      "eval_accuracy": 0.0812962962962963,
      "eval_loss": 5.572833061218262,
      "eval_runtime": 42.3306,
      "eval_samples_per_second": 127.567,
      "eval_steps_per_second": 15.946,
      "step": 2800
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.189086294416244e-05,
      "loss": 0.4932,
      "step": 2810
    },
    {
      "epoch": 3.58,
      "learning_rate": 2.1256345177664976e-05,
      "loss": 0.477,
      "step": 2820
    },
    {
      "epoch": 3.59,
      "learning_rate": 2.0621827411167513e-05,
      "loss": 0.4608,
      "step": 2830
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.9987309644670053e-05,
      "loss": 0.4357,
      "step": 2840
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.935279187817259e-05,
      "loss": 0.4522,
      "step": 2850
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.8718274111675127e-05,
      "loss": 0.4454,
      "step": 2860
    },
    {
      "epoch": 3.64,
      "learning_rate": 1.8083756345177667e-05,
      "loss": 0.5819,
      "step": 2870
    },
    {
      "epoch": 3.65,
      "learning_rate": 1.7449238578680204e-05,
      "loss": 0.4609,
      "step": 2880
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.681472081218274e-05,
      "loss": 0.4071,
      "step": 2890
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.618020304568528e-05,
      "loss": 0.5932,
      "step": 2900
    },
    {
      "epoch": 3.68,
      "eval_accuracy": 0.0662962962962963,
      "eval_loss": 5.799234867095947,
      "eval_runtime": 41.9548,
      "eval_samples_per_second": 128.71,
      "eval_steps_per_second": 16.089,
      "step": 2900
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.5545685279187817e-05,
      "loss": 0.427,
      "step": 2910
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.4911167512690354e-05,
      "loss": 0.5988,
      "step": 2920
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.4276649746192894e-05,
      "loss": 0.517,
      "step": 2930
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3642131979695433e-05,
      "loss": 0.4493,
      "step": 2940
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.3007614213197968e-05,
      "loss": 0.5024,
      "step": 2950
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.2373096446700508e-05,
      "loss": 0.6078,
      "step": 2960
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.1738578680203047e-05,
      "loss": 0.4253,
      "step": 2970
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.1104060913705584e-05,
      "loss": 0.3708,
      "step": 2980
    },
    {
      "epoch": 3.79,
      "learning_rate": 1.0469543147208122e-05,
      "loss": 0.4646,
      "step": 2990
    },
    {
      "epoch": 3.81,
      "learning_rate": 9.83502538071066e-06,
      "loss": 0.4382,
      "step": 3000
    },
    {
      "epoch": 3.81,
      "eval_accuracy": 0.06425925925925927,
      "eval_loss": 5.801009178161621,
      "eval_runtime": 42.6684,
      "eval_samples_per_second": 126.557,
      "eval_steps_per_second": 15.82,
      "step": 3000
    },
    {
      "epoch": 3.82,
      "learning_rate": 9.200507614213197e-06,
      "loss": 0.6405,
      "step": 3010
    },
    {
      "epoch": 3.83,
      "learning_rate": 8.565989847715736e-06,
      "loss": 0.4097,
      "step": 3020
    },
    {
      "epoch": 3.85,
      "learning_rate": 7.931472081218275e-06,
      "loss": 0.4651,
      "step": 3030
    },
    {
      "epoch": 3.86,
      "learning_rate": 7.296954314720812e-06,
      "loss": 0.7308,
      "step": 3040
    },
    {
      "epoch": 3.87,
      "learning_rate": 6.6624365482233515e-06,
      "loss": 0.4335,
      "step": 3050
    },
    {
      "epoch": 3.88,
      "learning_rate": 6.027918781725888e-06,
      "loss": 0.591,
      "step": 3060
    },
    {
      "epoch": 3.9,
      "learning_rate": 5.393401015228427e-06,
      "loss": 0.4654,
      "step": 3070
    },
    {
      "epoch": 3.91,
      "learning_rate": 4.7588832487309645e-06,
      "loss": 0.4933,
      "step": 3080
    },
    {
      "epoch": 3.92,
      "learning_rate": 4.124365482233502e-06,
      "loss": 0.5115,
      "step": 3090
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.4898477157360407e-06,
      "loss": 0.4827,
      "step": 3100
    },
    {
      "epoch": 3.93,
      "eval_accuracy": 0.06796296296296296,
      "eval_loss": 5.752891540527344,
      "eval_runtime": 42.606,
      "eval_samples_per_second": 126.743,
      "eval_steps_per_second": 15.843,
      "step": 3100
    },
    {
      "epoch": 3.95,
      "learning_rate": 2.855329949238579e-06,
      "loss": 0.4133,
      "step": 3110
    },
    {
      "epoch": 3.96,
      "learning_rate": 2.220812182741117e-06,
      "loss": 0.5857,
      "step": 3120
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.5862944162436548e-06,
      "loss": 0.3375,
      "step": 3130
    },
    {
      "epoch": 3.98,
      "learning_rate": 9.51776649746193e-07,
      "loss": 0.6126,
      "step": 3140
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.1725888324873095e-07,
      "loss": 0.5998,
      "step": 3150
    },
    {
      "epoch": 4.0,
      "step": 3152,
      "total_flos": 3.906051336688435e+18,
      "train_loss": 0.813563161695064,
      "train_runtime": 2380.4416,
      "train_samples_per_second": 21.173,
      "train_steps_per_second": 1.324
    }
  ],
  "max_steps": 3152,
  "num_train_epochs": 4,
  "total_flos": 3.906051336688435e+18,
  "trial_name": null,
  "trial_params": null
}