{
  "best_metric": 0.35135135135135137,
  "best_model_checkpoint": "vivit-b-16x2-kinetics400-finetuned-ucf101-subset-without-pretrained/checkpoint-2100",
  "epoch": 9.1,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 6.0146,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 5.6079,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-06,
      "loss": 5.4694,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.666666666666667e-06,
      "loss": 4.7757,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.333333333333334e-06,
      "loss": 4.2847,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 1e-05,
      "loss": 3.6053,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.1666666666666668e-05,
      "loss": 3.3222,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 4.0703,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5e-05,
      "loss": 2.4623,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 4.2615,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.8333333333333333e-05,
      "loss": 3.0162,
      "step": 110
    },
    {
      "epoch": 0.04,
      "learning_rate": 2e-05,
      "loss": 2.8954,
      "step": 120
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.1666666666666667e-05,
      "loss": 2.3622,
      "step": 130
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 2.9892,
      "step": 140
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.5e-05,
      "loss": 2.4459,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 2.2578,
      "step": 160
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.8333333333333335e-05,
      "loss": 3.7852,
      "step": 170
    },
    {
      "epoch": 0.06,
      "learning_rate": 3e-05,
      "loss": 3.472,
      "step": 180
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.1666666666666666e-05,
      "loss": 2.0967,
      "step": 190
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.8889,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.5e-05,
      "loss": 3.5945,
      "step": 210
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 2.4664,
      "step": 220
    },
    {
      "epoch": 0.08,
      "learning_rate": 3.8333333333333334e-05,
      "loss": 3.461,
      "step": 230
    },
    {
      "epoch": 0.08,
      "learning_rate": 4e-05,
      "loss": 3.1201,
      "step": 240
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.166666666666667e-05,
      "loss": 2.6063,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 1.8298,
      "step": 260
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.5e-05,
      "loss": 4.0171,
      "step": 270
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.666666666666667e-05,
      "loss": 2.9952,
      "step": 280
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.8333333333333334e-05,
      "loss": 2.5919,
      "step": 290
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-05,
      "loss": 3.2964,
      "step": 300
    },
    {
      "epoch": 0.1,
      "eval_accuracy": 0.10810810810810811,
      "eval_loss": 2.4748547077178955,
      "eval_runtime": 7.888,
      "eval_samples_per_second": 4.691,
      "eval_steps_per_second": 4.691,
      "step": 300
    },
    {
      "epoch": 1.0,
      "learning_rate": 4.981481481481482e-05,
      "loss": 2.1251,
      "step": 310
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.962962962962963e-05,
      "loss": 2.7957,
      "step": 320
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.9444444444444446e-05,
      "loss": 3.0016,
      "step": 330
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.925925925925926e-05,
      "loss": 2.9296,
      "step": 340
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.9074074074074075e-05,
      "loss": 2.5054,
      "step": 350
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.888888888888889e-05,
      "loss": 2.6099,
      "step": 360
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.8703703703703704e-05,
      "loss": 2.3462,
      "step": 370
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.851851851851852e-05,
      "loss": 2.5099,
      "step": 380
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.8333333333333334e-05,
      "loss": 2.9488,
      "step": 390
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.814814814814815e-05,
      "loss": 3.1838,
      "step": 400
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.796296296296296e-05,
      "loss": 2.2918,
      "step": 410
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.7777777777777784e-05,
      "loss": 2.622,
      "step": 420
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.759259259259259e-05,
      "loss": 2.1096,
      "step": 430
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.740740740740741e-05,
      "loss": 2.3935,
      "step": 440
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.722222222222222e-05,
      "loss": 2.2316,
      "step": 450
    },
    {
      "epoch": 1.05,
      "learning_rate": 4.703703703703704e-05,
      "loss": 2.476,
      "step": 460
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.685185185185185e-05,
      "loss": 2.5298,
      "step": 470
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.666666666666667e-05,
      "loss": 2.5783,
      "step": 480
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.648148148148148e-05,
      "loss": 2.4782,
      "step": 490
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.62962962962963e-05,
      "loss": 2.3204,
      "step": 500
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.6111111111111115e-05,
      "loss": 2.6603,
      "step": 510
    },
    {
      "epoch": 1.07,
      "learning_rate": 4.592592592592593e-05,
      "loss": 2.1841,
      "step": 520
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.5740740740740745e-05,
      "loss": 2.1368,
      "step": 530
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.555555555555556e-05,
      "loss": 2.2092,
      "step": 540
    },
    {
      "epoch": 1.08,
      "learning_rate": 4.5370370370370374e-05,
      "loss": 2.6375,
      "step": 550
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.518518518518519e-05,
      "loss": 2.0985,
      "step": 560
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.5e-05,
      "loss": 2.6354,
      "step": 570
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.481481481481482e-05,
      "loss": 2.0626,
      "step": 580
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.462962962962963e-05,
      "loss": 1.8691,
      "step": 590
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 3.1346,
      "step": 600
    },
    {
      "epoch": 1.1,
      "eval_accuracy": 0.10810810810810811,
      "eval_loss": 2.4779040813446045,
      "eval_runtime": 7.7884,
      "eval_samples_per_second": 4.751,
      "eval_steps_per_second": 4.751,
      "step": 600
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.425925925925926e-05,
      "loss": 2.8789,
      "step": 610
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.4074074074074076e-05,
      "loss": 2.1999,
      "step": 620
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.388888888888889e-05,
      "loss": 2.4515,
      "step": 630
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.3703703703703705e-05,
      "loss": 2.0012,
      "step": 640
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.351851851851852e-05,
      "loss": 2.0266,
      "step": 650
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 1.6561,
      "step": 660
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.314814814814815e-05,
      "loss": 2.2148,
      "step": 670
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.296296296296296e-05,
      "loss": 2.4418,
      "step": 680
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.277777777777778e-05,
      "loss": 2.9789,
      "step": 690
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.259259259259259e-05,
      "loss": 1.9881,
      "step": 700
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.240740740740741e-05,
      "loss": 2.2087,
      "step": 710
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.222222222222222e-05,
      "loss": 2.1964,
      "step": 720
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.203703703703704e-05,
      "loss": 2.5474,
      "step": 730
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.185185185185185e-05,
      "loss": 2.1304,
      "step": 740
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.8592,
      "step": 750
    },
    {
      "epoch": 2.05,
      "learning_rate": 4.148148148148148e-05,
      "loss": 2.8711,
      "step": 760
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.12962962962963e-05,
      "loss": 2.469,
      "step": 770
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.111111111111111e-05,
      "loss": 1.9158,
      "step": 780
    },
    {
      "epoch": 2.06,
      "learning_rate": 4.092592592592593e-05,
      "loss": 2.0509,
      "step": 790
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.074074074074074e-05,
      "loss": 1.7933,
      "step": 800
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.055555555555556e-05,
      "loss": 2.0508,
      "step": 810
    },
    {
      "epoch": 2.07,
      "learning_rate": 4.0370370370370374e-05,
      "loss": 2.1783,
      "step": 820
    },
    {
      "epoch": 2.08,
      "learning_rate": 4.018518518518519e-05,
      "loss": 2.4269,
      "step": 830
    },
    {
      "epoch": 2.08,
      "learning_rate": 4e-05,
      "loss": 2.5345,
      "step": 840
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.981481481481482e-05,
      "loss": 2.0431,
      "step": 850
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.962962962962963e-05,
      "loss": 2.3893,
      "step": 860
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.944444444444445e-05,
      "loss": 2.3061,
      "step": 870
    },
    {
      "epoch": 2.09,
      "learning_rate": 3.925925925925926e-05,
      "loss": 1.8541,
      "step": 880
    },
    {
      "epoch": 2.1,
      "learning_rate": 3.9074074074074076e-05,
      "loss": 2.2115,
      "step": 890
    },
    {
      "epoch": 2.1,
      "learning_rate": 3.888888888888889e-05,
      "loss": 1.9796,
      "step": 900
    },
    {
      "epoch": 2.1,
      "eval_accuracy": 0.2972972972972973,
      "eval_loss": 2.4338538646698,
      "eval_runtime": 7.9053,
      "eval_samples_per_second": 4.68,
      "eval_steps_per_second": 4.68,
      "step": 900
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.8703703703703705e-05,
      "loss": 1.8704,
      "step": 910
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.851851851851852e-05,
      "loss": 1.8479,
      "step": 920
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.8333333333333334e-05,
      "loss": 1.6709,
      "step": 930
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.814814814814815e-05,
      "loss": 1.9238,
      "step": 940
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.7962962962962964e-05,
      "loss": 2.2086,
      "step": 950
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.777777777777778e-05,
      "loss": 2.0035,
      "step": 960
    },
    {
      "epoch": 3.02,
      "learning_rate": 3.759259259259259e-05,
      "loss": 1.959,
      "step": 970
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.740740740740741e-05,
      "loss": 1.736,
      "step": 980
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.722222222222222e-05,
      "loss": 2.4448,
      "step": 990
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 2.1386,
      "step": 1000
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.685185185185185e-05,
      "loss": 2.0454,
      "step": 1010
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 2.6412,
      "step": 1020
    },
    {
      "epoch": 3.04,
      "learning_rate": 3.648148148148148e-05,
      "loss": 2.0792,
      "step": 1030
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.62962962962963e-05,
      "loss": 2.531,
      "step": 1040
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.611111111111111e-05,
      "loss": 1.9005,
      "step": 1050
    },
    {
      "epoch": 3.05,
      "learning_rate": 3.592592592592593e-05,
      "loss": 2.1969,
      "step": 1060
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.574074074074074e-05,
      "loss": 1.8413,
      "step": 1070
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.555555555555556e-05,
      "loss": 2.0399,
      "step": 1080
    },
    {
      "epoch": 3.06,
      "learning_rate": 3.537037037037037e-05,
      "loss": 1.9236,
      "step": 1090
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.518518518518519e-05,
      "loss": 1.4469,
      "step": 1100
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.5e-05,
      "loss": 1.9591,
      "step": 1110
    },
    {
      "epoch": 3.07,
      "learning_rate": 3.481481481481482e-05,
      "loss": 2.4234,
      "step": 1120
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.4629629629629626e-05,
      "loss": 2.0327,
      "step": 1130
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.444444444444445e-05,
      "loss": 2.8836,
      "step": 1140
    },
    {
      "epoch": 3.08,
      "learning_rate": 3.425925925925926e-05,
      "loss": 2.1492,
      "step": 1150
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.4074074074074077e-05,
      "loss": 1.9895,
      "step": 1160
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.388888888888889e-05,
      "loss": 2.1062,
      "step": 1170
    },
    {
      "epoch": 3.09,
      "learning_rate": 3.3703703703703706e-05,
      "loss": 2.2514,
      "step": 1180
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.351851851851852e-05,
      "loss": 1.5223,
      "step": 1190
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.8692,
      "step": 1200
    },
    {
      "epoch": 3.1,
      "eval_accuracy": 0.32432432432432434,
      "eval_loss": 2.177396535873413,
      "eval_runtime": 7.9365,
      "eval_samples_per_second": 4.662,
      "eval_steps_per_second": 4.662,
      "step": 1200
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.314814814814815e-05,
      "loss": 1.8137,
      "step": 1210
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.2962962962962964e-05,
      "loss": 1.8109,
      "step": 1220
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.277777777777778e-05,
      "loss": 1.4493,
      "step": 1230
    },
    {
      "epoch": 4.01,
      "learning_rate": 3.25925925925926e-05,
      "loss": 1.6365,
      "step": 1240
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.240740740740741e-05,
      "loss": 2.534,
      "step": 1250
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.222222222222223e-05,
      "loss": 2.5825,
      "step": 1260
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.203703703703704e-05,
      "loss": 2.208,
      "step": 1270
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.185185185185185e-05,
      "loss": 1.6943,
      "step": 1280
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.1666666666666666e-05,
      "loss": 1.9221,
      "step": 1290
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.148148148148148e-05,
      "loss": 1.3294,
      "step": 1300
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.1296296296296295e-05,
      "loss": 2.6418,
      "step": 1310
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.111111111111111e-05,
      "loss": 2.0143,
      "step": 1320
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.0925925925925924e-05,
      "loss": 1.5268,
      "step": 1330
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.074074074074074e-05,
      "loss": 2.2358,
      "step": 1340
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.055555555555556e-05,
      "loss": 2.1382,
      "step": 1350
    },
    {
      "epoch": 4.05,
      "learning_rate": 3.037037037037037e-05,
      "loss": 2.4677,
      "step": 1360
    },
    {
      "epoch": 4.06,
      "learning_rate": 3.018518518518519e-05,
      "loss": 1.0908,
      "step": 1370
    },
    {
      "epoch": 4.06,
      "learning_rate": 3e-05,
      "loss": 2.109,
      "step": 1380
    },
    {
      "epoch": 4.06,
      "learning_rate": 2.981481481481482e-05,
      "loss": 3.2783,
      "step": 1390
    },
    {
      "epoch": 4.07,
      "learning_rate": 2.962962962962963e-05,
      "loss": 1.6895,
      "step": 1400
    },
    {
      "epoch": 4.07,
      "learning_rate": 2.9444444444444448e-05,
      "loss": 2.4235,
      "step": 1410
    },
    {
      "epoch": 4.07,
      "learning_rate": 2.925925925925926e-05,
      "loss": 2.7382,
      "step": 1420
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.9074074074074077e-05,
      "loss": 2.0283,
      "step": 1430
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 2.1473,
      "step": 1440
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.8703703703703706e-05,
      "loss": 2.0384,
      "step": 1450
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.851851851851852e-05,
      "loss": 1.7009,
      "step": 1460
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.8333333333333335e-05,
      "loss": 1.6367,
      "step": 1470
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.814814814814815e-05,
      "loss": 1.9183,
      "step": 1480
    },
    {
      "epoch": 4.1,
      "learning_rate": 2.7962962962962965e-05,
      "loss": 2.9585,
      "step": 1490
    },
    {
      "epoch": 4.1,
      "learning_rate": 2.777777777777778e-05,
      "loss": 1.3617,
      "step": 1500
    },
    {
      "epoch": 4.1,
      "eval_accuracy": 0.32432432432432434,
      "eval_loss": 2.161275625228882,
      "eval_runtime": 7.8706,
      "eval_samples_per_second": 4.701,
      "eval_steps_per_second": 4.701,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "learning_rate": 2.7592592592592594e-05,
      "loss": 2.0505,
      "step": 1510
    },
    {
      "epoch": 5.01,
      "learning_rate": 2.7407407407407408e-05,
      "loss": 1.6757,
      "step": 1520
    },
    {
      "epoch": 5.01,
      "learning_rate": 2.7222222222222223e-05,
      "loss": 1.791,
      "step": 1530
    },
    {
      "epoch": 5.01,
      "learning_rate": 2.7037037037037037e-05,
      "loss": 1.8723,
      "step": 1540
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.6851851851851855e-05,
      "loss": 2.2865,
      "step": 1550
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 1.5064,
      "step": 1560
    },
    {
      "epoch": 5.02,
      "learning_rate": 2.6481481481481485e-05,
      "loss": 1.462,
      "step": 1570
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.6296296296296296e-05,
      "loss": 1.7543,
      "step": 1580
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.6111111111111114e-05,
      "loss": 1.2975,
      "step": 1590
    },
    {
      "epoch": 5.03,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 2.18,
      "step": 1600
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.5740740740740743e-05,
      "loss": 2.3015,
      "step": 1610
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.5555555555555554e-05,
      "loss": 1.0242,
      "step": 1620
    },
    {
      "epoch": 5.04,
      "learning_rate": 2.5370370370370372e-05,
      "loss": 1.6023,
      "step": 1630
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.5185185185185183e-05,
      "loss": 1.9401,
      "step": 1640
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.5e-05,
      "loss": 1.413,
      "step": 1650
    },
    {
      "epoch": 5.05,
      "learning_rate": 2.4814814814814816e-05,
      "loss": 2.5201,
      "step": 1660
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.462962962962963e-05,
      "loss": 2.3386,
      "step": 1670
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 2.4826,
      "step": 1680
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.425925925925926e-05,
      "loss": 1.8664,
      "step": 1690
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 1.6916,
      "step": 1700
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.3888888888888892e-05,
      "loss": 1.4313,
      "step": 1710
    },
    {
      "epoch": 5.07,
      "learning_rate": 2.3703703703703707e-05,
      "loss": 2.0346,
      "step": 1720
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.351851851851852e-05,
      "loss": 1.9452,
      "step": 1730
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 1.414,
      "step": 1740
    },
    {
      "epoch": 5.08,
      "learning_rate": 2.314814814814815e-05,
      "loss": 1.6949,
      "step": 1750
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.2962962962962965e-05,
      "loss": 2.2121,
      "step": 1760
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.277777777777778e-05,
      "loss": 2.301,
      "step": 1770
    },
    {
      "epoch": 5.09,
      "learning_rate": 2.2592592592592594e-05,
      "loss": 2.1127,
      "step": 1780
    },
    {
      "epoch": 5.1,
      "learning_rate": 2.240740740740741e-05,
      "loss": 1.931,
      "step": 1790
    },
    {
      "epoch": 5.1,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 2.0821,
      "step": 1800
    },
    {
      "epoch": 5.1,
      "eval_accuracy": 0.21621621621621623,
      "eval_loss": 1.9206827878952026,
      "eval_runtime": 7.9165,
      "eval_samples_per_second": 4.674,
      "eval_steps_per_second": 4.674,
      "step": 1800
    },
    {
      "epoch": 6.0,
      "learning_rate": 2.2037037037037038e-05,
      "loss": 1.7858,
      "step": 1810
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.1851851851851852e-05,
      "loss": 1.8683,
      "step": 1820
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.1666666666666667e-05,
      "loss": 1.6547,
      "step": 1830
    },
    {
      "epoch": 6.01,
      "learning_rate": 2.148148148148148e-05,
      "loss": 0.798,
      "step": 1840
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.1296296296296296e-05,
      "loss": 3.1924,
      "step": 1850
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.111111111111111e-05,
      "loss": 3.1257,
      "step": 1860
    },
    {
      "epoch": 6.02,
      "learning_rate": 2.0925925925925925e-05,
      "loss": 1.8917,
      "step": 1870
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.074074074074074e-05,
      "loss": 1.0172,
      "step": 1880
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.0555555555555555e-05,
      "loss": 2.2976,
      "step": 1890
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.037037037037037e-05,
      "loss": 1.7965,
      "step": 1900
    },
    {
      "epoch": 6.04,
      "learning_rate": 2.0185185185185187e-05,
      "loss": 1.7191,
      "step": 1910
    },
    {
      "epoch": 6.04,
      "learning_rate": 2e-05,
      "loss": 1.9343,
      "step": 1920
    },
    {
      "epoch": 6.04,
      "learning_rate": 1.9814814814814816e-05,
      "loss": 2.0969,
      "step": 1930
    },
    {
      "epoch": 6.05,
      "learning_rate": 1.962962962962963e-05,
      "loss": 1.1947,
      "step": 1940
    },
    {
      "epoch": 6.05,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 1.1954,
      "step": 1950
    },
    {
      "epoch": 6.05,
      "learning_rate": 1.925925925925926e-05,
      "loss": 1.9325,
      "step": 1960
    },
    {
      "epoch": 6.06,
      "learning_rate": 1.9074074074074075e-05,
      "loss": 1.4172,
      "step": 1970
    },
    {
      "epoch": 6.06,
      "learning_rate": 1.888888888888889e-05,
      "loss": 1.6606,
      "step": 1980
    },
    {
      "epoch": 6.06,
      "learning_rate": 1.8703703703703704e-05,
      "loss": 0.719,
      "step": 1990
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 1.7754,
      "step": 2000
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.8333333333333333e-05,
      "loss": 1.5815,
      "step": 2010
    },
    {
      "epoch": 6.07,
      "learning_rate": 1.814814814814815e-05,
      "loss": 2.0108,
      "step": 2020
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.7962962962962965e-05,
      "loss": 0.4663,
      "step": 2030
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.777777777777778e-05,
      "loss": 2.8609,
      "step": 2040
    },
    {
      "epoch": 6.08,
      "learning_rate": 1.7592592592592595e-05,
      "loss": 1.6084,
      "step": 2050
    },
    {
      "epoch": 6.09,
      "learning_rate": 1.740740740740741e-05,
      "loss": 1.6349,
      "step": 2060
    },
    {
      "epoch": 6.09,
      "learning_rate": 1.7222222222222224e-05,
      "loss": 2.4036,
      "step": 2070
    },
    {
      "epoch": 6.09,
      "learning_rate": 1.7037037037037038e-05,
      "loss": 2.9006,
      "step": 2080
    },
    {
      "epoch": 6.1,
      "learning_rate": 1.6851851851851853e-05,
      "loss": 1.9097,
      "step": 2090
    },
    {
      "epoch": 6.1,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.5534,
      "step": 2100
    },
    {
      "epoch": 6.1,
      "eval_accuracy": 0.35135135135135137,
      "eval_loss": 2.200272560119629,
      "eval_runtime": 8.1085,
      "eval_samples_per_second": 4.563,
      "eval_steps_per_second": 4.563,
      "step": 2100
    },
    {
      "epoch": 7.0,
      "learning_rate": 1.6481481481481482e-05,
      "loss": 1.1439,
      "step": 2110
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.62962962962963e-05,
      "loss": 1.6308,
      "step": 2120
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.6111111111111115e-05,
      "loss": 1.7053,
      "step": 2130
    },
    {
      "epoch": 7.01,
      "learning_rate": 1.5925925925925926e-05,
      "loss": 1.1133,
      "step": 2140
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.574074074074074e-05,
      "loss": 3.4859,
      "step": 2150
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 1.781,
      "step": 2160
    },
    {
      "epoch": 7.02,
      "learning_rate": 1.537037037037037e-05,
      "loss": 1.0939,
      "step": 2170
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.5185185185185186e-05,
      "loss": 1.2528,
      "step": 2180
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.5e-05,
      "loss": 1.2228,
      "step": 2190
    },
    {
      "epoch": 7.03,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 3.4645,
      "step": 2200
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.462962962962963e-05,
      "loss": 1.4841,
      "step": 2210
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.4444444444444444e-05,
      "loss": 1.4566,
      "step": 2220
    },
    {
      "epoch": 7.04,
      "learning_rate": 1.425925925925926e-05,
      "loss": 1.5839,
      "step": 2230
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.4074074074074075e-05,
      "loss": 0.9707,
      "step": 2240
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.388888888888889e-05,
      "loss": 3.4078,
      "step": 2250
    },
    {
      "epoch": 7.05,
      "learning_rate": 1.3703703703703704e-05,
      "loss": 2.6832,
      "step": 2260
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.3518518518518519e-05,
      "loss": 1.4994,
      "step": 2270
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 1.5557,
      "step": 2280
    },
    {
      "epoch": 7.06,
      "learning_rate": 1.3148148148148148e-05,
      "loss": 1.4492,
      "step": 2290
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 1.345,
      "step": 2300
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.2777777777777777e-05,
      "loss": 1.5474,
      "step": 2310
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.2592592592592592e-05,
      "loss": 2.1177,
      "step": 2320
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.2407407407407408e-05,
      "loss": 1.4745,
      "step": 2330
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.2222222222222222e-05,
      "loss": 1.0913,
      "step": 2340
    },
    {
      "epoch": 7.08,
      "learning_rate": 1.2037037037037037e-05,
      "loss": 3.3812,
      "step": 2350
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.1851851851851853e-05,
      "loss": 1.1772,
      "step": 2360
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.1666666666666668e-05,
      "loss": 1.407,
      "step": 2370
    },
    {
      "epoch": 7.09,
      "learning_rate": 1.1481481481481482e-05,
      "loss": 1.1857,
      "step": 2380
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.1296296296296297e-05,
      "loss": 1.3831,
      "step": 2390
    },
    {
      "epoch": 7.1,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.5632,
      "step": 2400
    },
    {
      "epoch": 7.1,
      "eval_accuracy": 0.35135135135135137,
      "eval_loss": 3.021118402481079,
      "eval_runtime": 7.9126,
      "eval_samples_per_second": 4.676,
      "eval_steps_per_second": 4.676,
      "step": 2400
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.0925925925925926e-05,
      "loss": 1.528,
      "step": 2410
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.074074074074074e-05,
      "loss": 1.5443,
      "step": 2420
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.0555555555555555e-05,
      "loss": 2.7518,
      "step": 2430
    },
    {
      "epoch": 8.01,
      "learning_rate": 1.037037037037037e-05,
      "loss": 1.5082,
      "step": 2440
    },
    {
      "epoch": 8.02,
      "learning_rate": 1.0185185185185185e-05,
      "loss": 2.108,
      "step": 2450
    },
    {
      "epoch": 8.02,
      "learning_rate": 1e-05,
      "loss": 1.9699,
      "step": 2460
    },
    {
      "epoch": 8.02,
      "learning_rate": 9.814814814814815e-06,
      "loss": 1.2374,
      "step": 2470
    },
    {
      "epoch": 8.03,
      "learning_rate": 9.62962962962963e-06,
      "loss": 2.103,
      "step": 2480
    },
    {
      "epoch": 8.03,
      "learning_rate": 9.444444444444445e-06,
      "loss": 0.9801,
      "step": 2490
    },
    {
      "epoch": 8.03,
      "learning_rate": 9.259259259259259e-06,
      "loss": 2.5852,
      "step": 2500
    },
    {
      "epoch": 8.04,
      "learning_rate": 9.074074074074075e-06,
      "loss": 1.9504,
      "step": 2510
    },
    {
      "epoch": 8.04,
      "learning_rate": 8.88888888888889e-06,
      "loss": 2.4488,
      "step": 2520
    },
    {
      "epoch": 8.04,
      "learning_rate": 8.703703703703705e-06,
      "loss": 0.9139,
      "step": 2530
    },
    {
      "epoch": 8.05,
      "learning_rate": 8.518518518518519e-06,
      "loss": 0.995,
      "step": 2540
    },
    {
      "epoch": 8.05,
      "learning_rate": 8.333333333333334e-06,
      "loss": 1.7461,
      "step": 2550
    },
    {
      "epoch": 8.05,
      "learning_rate": 8.14814814814815e-06,
      "loss": 1.1371,
      "step": 2560
    },
    {
      "epoch": 8.06,
      "learning_rate": 7.962962962962963e-06,
      "loss": 0.9142,
      "step": 2570
    },
    {
      "epoch": 8.06,
      "learning_rate": 7.777777777777777e-06,
      "loss": 1.9478,
      "step": 2580
    },
    {
      "epoch": 8.06,
      "learning_rate": 7.592592592592593e-06,
      "loss": 2.1475,
      "step": 2590
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 1.991,
      "step": 2600
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.222222222222222e-06,
      "loss": 1.4008,
      "step": 2610
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.0370370370370375e-06,
      "loss": 1.2762,
      "step": 2620
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.851851851851852e-06,
      "loss": 1.3275,
      "step": 2630
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.6279,
      "step": 2640
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.481481481481481e-06,
      "loss": 1.293,
      "step": 2650
    },
    {
      "epoch": 8.09,
      "learning_rate": 6.296296296296296e-06,
      "loss": 0.754,
      "step": 2660
    },
    {
      "epoch": 8.09,
      "learning_rate": 6.111111111111111e-06,
      "loss": 1.514,
      "step": 2670
    },
    {
      "epoch": 8.09,
      "learning_rate": 5.925925925925927e-06,
      "loss": 1.885,
      "step": 2680
    },
    {
      "epoch": 8.1,
      "learning_rate": 5.740740740740741e-06,
      "loss": 1.7632,
      "step": 2690
    },
    {
      "epoch": 8.1,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.2806,
      "step": 2700
    },
    {
      "epoch": 8.1,
      "eval_accuracy": 0.35135135135135137,
      "eval_loss": 2.7735354900360107,
      "eval_runtime": 8.0014,
      "eval_samples_per_second": 4.624,
      "eval_steps_per_second": 4.624,
      "step": 2700
    },
    {
      "epoch": 9.0,
      "learning_rate": 5.37037037037037e-06,
      "loss": 1.1323,
      "step": 2710
    },
    {
      "epoch": 9.01,
      "learning_rate": 5.185185185185185e-06,
      "loss": 0.3361,
      "step": 2720
    },
    {
      "epoch": 9.01,
      "learning_rate": 5e-06,
      "loss": 1.5187,
      "step": 2730
    },
    {
      "epoch": 9.01,
      "learning_rate": 4.814814814814815e-06,
      "loss": 1.0224,
      "step": 2740
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.9034,
      "step": 2750
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.444444444444445e-06,
      "loss": 1.0158,
      "step": 2760
    },
    {
      "epoch": 9.02,
      "learning_rate": 4.2592592592592596e-06,
      "loss": 0.6375,
      "step": 2770
    },
    {
      "epoch": 9.03,
      "learning_rate": 4.074074074074075e-06,
      "loss": 1.8772,
      "step": 2780
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.888888888888889e-06,
      "loss": 1.0748,
      "step": 2790
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.9857,
      "step": 2800
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.5185185185185187e-06,
      "loss": 2.9974,
      "step": 2810
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.9478,
      "step": 2820
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.148148148148148e-06,
      "loss": 1.3206,
      "step": 2830
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.9629629629629633e-06,
      "loss": 2.1553,
      "step": 2840
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.777777777777778e-06,
      "loss": 2.3575,
      "step": 2850
    },
    {
      "epoch": 9.05,
      "learning_rate": 2.5925925925925925e-06,
      "loss": 1.433,
      "step": 2860
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.4074074074074075e-06,
      "loss": 0.9617,
      "step": 2870
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.2222222222222225e-06,
      "loss": 1.1871,
      "step": 2880
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.0370370370370375e-06,
      "loss": 1.5125,
      "step": 2890
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.8774,
      "step": 2900
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 2.0533,
      "step": 2910
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.4814814814814817e-06,
      "loss": 1.3807,
      "step": 2920
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.2962962962962962e-06,
      "loss": 1.394,
      "step": 2930
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.1111111111111112e-06,
      "loss": 1.5122,
      "step": 2940
    },
    {
      "epoch": 9.08,
      "learning_rate": 9.259259259259259e-07,
      "loss": 0.8405,
      "step": 2950
    },
    {
      "epoch": 9.09,
      "learning_rate": 7.407407407407408e-07,
      "loss": 0.7064,
      "step": 2960
    },
    {
      "epoch": 9.09,
      "learning_rate": 5.555555555555556e-07,
      "loss": 0.9997,
      "step": 2970
    },
    {
      "epoch": 9.09,
      "learning_rate": 3.703703703703704e-07,
      "loss": 0.7745,
      "step": 2980
    },
    {
      "epoch": 9.1,
      "learning_rate": 1.851851851851852e-07,
      "loss": 1.0899,
      "step": 2990
    },
    {
      "epoch": 9.1,
      "learning_rate": 0.0,
      "loss": 1.3832,
      "step": 3000
    },
    {
      "epoch": 9.1,
      "eval_accuracy": 0.35135135135135137,
      "eval_loss": 2.931495189666748,
      "eval_runtime": 8.0466,
      "eval_samples_per_second": 4.598,
      "eval_steps_per_second": 4.598,
      "step": 3000
    },
    {
      "epoch": 9.1,
      "step": 3000,
      "total_flos": 7.712679002112e+18,
      "train_loss": 2.043955718199412,
      "train_runtime": 1387.0202,
      "train_samples_per_second": 2.163,
      "train_steps_per_second": 2.163
    },
    {
      "epoch": 9.1,
      "eval_accuracy": 0.41379310344827586,
      "eval_loss": 2.1643764972686768,
      "eval_runtime": 18.3497,
      "eval_samples_per_second": 4.741,
      "eval_steps_per_second": 4.741,
      "step": 3000
    },
    {
      "epoch": 9.1,
      "eval_accuracy": 0.41379310344827586,
      "eval_loss": 2.1643762588500977,
      "eval_runtime": 18.2718,
      "eval_samples_per_second": 4.761,
      "eval_steps_per_second": 4.761,
      "step": 3000
    }
  ],
  "logging_steps": 10,
  "max_steps": 3000,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 7.712679002112e+18,
  "trial_name": null,
  "trial_params": null
}