{
  "best_metric": 0.9285714285714286,
  "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-1200",
  "epoch": 3.25,
  "eval_steps": 500,
  "global_step": 1200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 39.893043518066406,
      "learning_rate": 4.166666666666667e-06,
      "loss": 2.3912,
      "step": 10
    },
    {
      "epoch": 0.02,
      "grad_norm": 21.392333984375,
      "learning_rate": 8.333333333333334e-06,
      "loss": 2.4635,
      "step": 20
    },
    {
      "epoch": 0.03,
      "grad_norm": 22.904409408569336,
      "learning_rate": 1.25e-05,
      "loss": 2.3054,
      "step": 30
    },
    {
      "epoch": 0.03,
      "grad_norm": 18.382219314575195,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 2.391,
      "step": 40
    },
    {
      "epoch": 0.04,
      "grad_norm": 18.44032859802246,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 2.3256,
      "step": 50
    },
    {
      "epoch": 0.05,
      "grad_norm": 20.57721519470215,
      "learning_rate": 2.5e-05,
      "loss": 2.2661,
      "step": 60
    },
    {
      "epoch": 0.06,
      "grad_norm": 27.851577758789062,
      "learning_rate": 2.916666666666667e-05,
      "loss": 2.252,
      "step": 70
    },
    {
      "epoch": 0.07,
      "grad_norm": 20.621288299560547,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.3036,
      "step": 80
    },
    {
      "epoch": 0.07,
      "grad_norm": 22.28342056274414,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 2.0055,
      "step": 90
    },
    {
      "epoch": 0.08,
      "grad_norm": 24.50592613220215,
      "learning_rate": 4.166666666666667e-05,
      "loss": 2.0529,
      "step": 100
    },
    {
      "epoch": 0.09,
      "grad_norm": 28.007957458496094,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 2.122,
      "step": 110
    },
    {
      "epoch": 0.1,
      "grad_norm": 22.624725341796875,
      "learning_rate": 5e-05,
      "loss": 2.1029,
      "step": 120
    },
    {
      "epoch": 0.11,
      "grad_norm": 23.730955123901367,
      "learning_rate": 4.9537037037037035e-05,
      "loss": 2.1566,
      "step": 130
    },
    {
      "epoch": 0.12,
      "grad_norm": 32.912437438964844,
      "learning_rate": 4.9074074074074075e-05,
      "loss": 1.9197,
      "step": 140
    },
    {
      "epoch": 0.12,
      "grad_norm": 28.358497619628906,
      "learning_rate": 4.8611111111111115e-05,
      "loss": 1.766,
      "step": 150
    },
    {
      "epoch": 0.13,
      "grad_norm": 38.748905181884766,
      "learning_rate": 4.814814814814815e-05,
      "loss": 1.5543,
      "step": 160
    },
    {
      "epoch": 0.14,
      "grad_norm": 19.396177291870117,
      "learning_rate": 4.768518518518519e-05,
      "loss": 1.8231,
      "step": 170
    },
    {
      "epoch": 0.15,
      "grad_norm": 20.669483184814453,
      "learning_rate": 4.722222222222222e-05,
      "loss": 1.8356,
      "step": 180
    },
    {
      "epoch": 0.16,
      "grad_norm": 16.87503433227539,
      "learning_rate": 4.675925925925926e-05,
      "loss": 2.0229,
      "step": 190
    },
    {
      "epoch": 0.17,
      "grad_norm": 37.929744720458984,
      "learning_rate": 4.62962962962963e-05,
      "loss": 1.4805,
      "step": 200
    },
    {
      "epoch": 0.17,
      "grad_norm": 23.615205764770508,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 2.5876,
      "step": 210
    },
    {
      "epoch": 0.18,
      "grad_norm": 11.07725715637207,
      "learning_rate": 4.5370370370370374e-05,
      "loss": 1.4919,
      "step": 220
    },
    {
      "epoch": 0.19,
      "grad_norm": 22.73790740966797,
      "learning_rate": 4.490740740740741e-05,
      "loss": 1.9309,
      "step": 230
    },
    {
      "epoch": 0.2,
      "grad_norm": 22.745410919189453,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 2.5053,
      "step": 240
    },
    {
      "epoch": 0.21,
      "grad_norm": 36.528160095214844,
      "learning_rate": 4.3981481481481486e-05,
      "loss": 2.1563,
      "step": 250
    },
    {
      "epoch": 0.22,
      "grad_norm": 64.73149108886719,
      "learning_rate": 4.351851851851852e-05,
      "loss": 2.5058,
      "step": 260
    },
    {
      "epoch": 0.23,
      "grad_norm": 33.93071365356445,
      "learning_rate": 4.305555555555556e-05,
      "loss": 2.1499,
      "step": 270
    },
    {
      "epoch": 0.23,
      "grad_norm": 23.45170783996582,
      "learning_rate": 4.259259259259259e-05,
      "loss": 1.7521,
      "step": 280
    },
    {
      "epoch": 0.24,
      "grad_norm": 21.1306209564209,
      "learning_rate": 4.212962962962963e-05,
      "loss": 1.1103,
      "step": 290
    },
    {
      "epoch": 0.25,
      "grad_norm": 75.54643249511719,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.7533,
      "step": 300
    },
    {
      "epoch": 0.25,
      "eval_accuracy": 0.32857142857142857,
      "eval_loss": 1.7997785806655884,
      "eval_runtime": 20.0261,
      "eval_samples_per_second": 3.495,
      "eval_steps_per_second": 3.495,
      "step": 300
    },
    {
      "epoch": 1.01,
      "grad_norm": 27.954673767089844,
      "learning_rate": 4.1203703703703705e-05,
      "loss": 1.0294,
      "step": 310
    },
    {
      "epoch": 1.02,
      "grad_norm": 29.380338668823242,
      "learning_rate": 4.074074074074074e-05,
      "loss": 1.5711,
      "step": 320
    },
    {
      "epoch": 1.02,
      "grad_norm": 105.077392578125,
      "learning_rate": 4.027777777777778e-05,
      "loss": 1.7588,
      "step": 330
    },
    {
      "epoch": 1.03,
      "grad_norm": 5.28167200088501,
      "learning_rate": 3.981481481481482e-05,
      "loss": 1.3781,
      "step": 340
    },
    {
      "epoch": 1.04,
      "grad_norm": 82.8572006225586,
      "learning_rate": 3.935185185185186e-05,
      "loss": 1.9861,
      "step": 350
    },
    {
      "epoch": 1.05,
      "grad_norm": 23.33498191833496,
      "learning_rate": 3.888888888888889e-05,
      "loss": 1.5307,
      "step": 360
    },
    {
      "epoch": 1.06,
      "grad_norm": 45.165435791015625,
      "learning_rate": 3.8425925925925924e-05,
      "loss": 0.9492,
      "step": 370
    },
    {
      "epoch": 1.07,
      "grad_norm": 1.3398464918136597,
      "learning_rate": 3.7962962962962964e-05,
      "loss": 1.3926,
      "step": 380
    },
    {
      "epoch": 1.07,
      "grad_norm": 29.025249481201172,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.7036,
      "step": 390
    },
    {
      "epoch": 1.08,
      "grad_norm": 42.30586624145508,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 1.1419,
      "step": 400
    },
    {
      "epoch": 1.09,
      "grad_norm": 23.552091598510742,
      "learning_rate": 3.6574074074074076e-05,
      "loss": 1.0914,
      "step": 410
    },
    {
      "epoch": 1.1,
      "grad_norm": 8.62332820892334,
      "learning_rate": 3.611111111111111e-05,
      "loss": 1.2866,
      "step": 420
    },
    {
      "epoch": 1.11,
      "grad_norm": 29.133724212646484,
      "learning_rate": 3.564814814814815e-05,
      "loss": 0.9732,
      "step": 430
    },
    {
      "epoch": 1.12,
      "grad_norm": 8.369429588317871,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.7792,
      "step": 440
    },
    {
      "epoch": 1.12,
      "grad_norm": 28.84503936767578,
      "learning_rate": 3.472222222222222e-05,
      "loss": 1.3297,
      "step": 450
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.6899290680885315,
      "learning_rate": 3.425925925925926e-05,
      "loss": 1.1334,
      "step": 460
    },
    {
      "epoch": 1.14,
      "grad_norm": 3.0040957927703857,
      "learning_rate": 3.3796296296296295e-05,
      "loss": 0.6843,
      "step": 470
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.4193795621395111,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.3539,
      "step": 480
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.723945140838623,
      "learning_rate": 3.2870370370370375e-05,
      "loss": 1.2607,
      "step": 490
    },
    {
      "epoch": 1.17,
      "grad_norm": 25.487699508666992,
      "learning_rate": 3.240740740740741e-05,
      "loss": 0.8374,
      "step": 500
    },
    {
      "epoch": 1.18,
      "grad_norm": 15.434425354003906,
      "learning_rate": 3.194444444444444e-05,
      "loss": 1.2351,
      "step": 510
    },
    {
      "epoch": 1.18,
      "grad_norm": 64.37287902832031,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.4603,
      "step": 520
    },
    {
      "epoch": 1.19,
      "grad_norm": 2.10884952545166,
      "learning_rate": 3.101851851851852e-05,
      "loss": 0.569,
      "step": 530
    },
    {
      "epoch": 1.2,
      "grad_norm": 38.082462310791016,
      "learning_rate": 3.055555555555556e-05,
      "loss": 2.0602,
      "step": 540
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.26877835392951965,
      "learning_rate": 3.0092592592592593e-05,
      "loss": 0.2031,
      "step": 550
    },
    {
      "epoch": 1.22,
      "grad_norm": 45.29481506347656,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.1889,
      "step": 560
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.8152311444282532,
      "learning_rate": 2.916666666666667e-05,
      "loss": 1.3687,
      "step": 570
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.16306141018867493,
      "learning_rate": 2.8703703703703706e-05,
      "loss": 1.4408,
      "step": 580
    },
    {
      "epoch": 1.24,
      "grad_norm": 29.856517791748047,
      "learning_rate": 2.824074074074074e-05,
      "loss": 1.8685,
      "step": 590
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.069265365600586,
      "learning_rate": 2.777777777777778e-05,
      "loss": 1.6337,
      "step": 600
    },
    {
      "epoch": 1.25,
      "eval_accuracy": 0.5571428571428572,
      "eval_loss": 1.2139314413070679,
      "eval_runtime": 19.6709,
      "eval_samples_per_second": 3.559,
      "eval_steps_per_second": 3.559,
      "step": 600
    },
    {
      "epoch": 2.01,
      "grad_norm": 53.667381286621094,
      "learning_rate": 2.7314814814814816e-05,
      "loss": 0.7485,
      "step": 610
    },
    {
      "epoch": 2.02,
      "grad_norm": 9.039528846740723,
      "learning_rate": 2.6851851851851855e-05,
      "loss": 0.5022,
      "step": 620
    },
    {
      "epoch": 2.02,
      "grad_norm": 2.096543550491333,
      "learning_rate": 2.6388888888888892e-05,
      "loss": 0.0704,
      "step": 630
    },
    {
      "epoch": 2.03,
      "grad_norm": 0.14604328572750092,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.6426,
      "step": 640
    },
    {
      "epoch": 2.04,
      "grad_norm": 80.3462905883789,
      "learning_rate": 2.5462962962962965e-05,
      "loss": 0.4556,
      "step": 650
    },
    {
      "epoch": 2.05,
      "grad_norm": 29.6263484954834,
      "learning_rate": 2.5e-05,
      "loss": 0.5818,
      "step": 660
    },
    {
      "epoch": 2.06,
      "grad_norm": 5.271122932434082,
      "learning_rate": 2.4537037037037038e-05,
      "loss": 0.2213,
      "step": 670
    },
    {
      "epoch": 2.07,
      "grad_norm": 37.88823699951172,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.7075,
      "step": 680
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.3303954005241394,
      "learning_rate": 2.361111111111111e-05,
      "loss": 0.3398,
      "step": 690
    },
    {
      "epoch": 2.08,
      "grad_norm": 7.315389633178711,
      "learning_rate": 2.314814814814815e-05,
      "loss": 0.3905,
      "step": 700
    },
    {
      "epoch": 2.09,
      "grad_norm": 5.788703441619873,
      "learning_rate": 2.2685185185185187e-05,
      "loss": 0.6432,
      "step": 710
    },
    {
      "epoch": 2.1,
      "grad_norm": 10.666046142578125,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1781,
      "step": 720
    },
    {
      "epoch": 2.11,
      "grad_norm": 0.13134099543094635,
      "learning_rate": 2.175925925925926e-05,
      "loss": 0.7766,
      "step": 730
    },
    {
      "epoch": 2.12,
      "grad_norm": 5.6078314781188965,
      "learning_rate": 2.1296296296296296e-05,
      "loss": 0.0573,
      "step": 740
    },
    {
      "epoch": 2.12,
      "grad_norm": 57.8688850402832,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.9231,
      "step": 750
    },
    {
      "epoch": 2.13,
      "grad_norm": 1.2845256328582764,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1482,
      "step": 760
    },
    {
      "epoch": 2.14,
      "grad_norm": 75.59434509277344,
      "learning_rate": 1.990740740740741e-05,
      "loss": 0.7703,
      "step": 770
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.07224389910697937,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.2001,
      "step": 780
    },
    {
      "epoch": 2.16,
      "grad_norm": 2.151550769805908,
      "learning_rate": 1.8981481481481482e-05,
      "loss": 0.6879,
      "step": 790
    },
    {
      "epoch": 2.17,
      "grad_norm": 0.5874713063240051,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.4532,
      "step": 800
    },
    {
      "epoch": 2.17,
      "grad_norm": 0.1461566686630249,
      "learning_rate": 1.8055555555555555e-05,
      "loss": 0.2781,
      "step": 810
    },
    {
      "epoch": 2.18,
      "grad_norm": 1.0537546873092651,
      "learning_rate": 1.7592592592592595e-05,
      "loss": 0.3174,
      "step": 820
    },
    {
      "epoch": 2.19,
      "grad_norm": 52.02813720703125,
      "learning_rate": 1.712962962962963e-05,
      "loss": 0.3687,
      "step": 830
    },
    {
      "epoch": 2.2,
      "grad_norm": 0.14705657958984375,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.3095,
      "step": 840
    },
    {
      "epoch": 2.21,
      "grad_norm": 10.978480339050293,
      "learning_rate": 1.6203703703703704e-05,
      "loss": 0.3199,
      "step": 850
    },
    {
      "epoch": 2.22,
      "grad_norm": 0.08446181565523148,
      "learning_rate": 1.574074074074074e-05,
      "loss": 0.1921,
      "step": 860
    },
    {
      "epoch": 2.23,
      "grad_norm": 0.10818950086832047,
      "learning_rate": 1.527777777777778e-05,
      "loss": 0.1006,
      "step": 870
    },
    {
      "epoch": 2.23,
      "grad_norm": 0.05260080099105835,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.0124,
      "step": 880
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.15152961015701294,
      "learning_rate": 1.4351851851851853e-05,
      "loss": 0.1896,
      "step": 890
    },
    {
      "epoch": 2.25,
      "grad_norm": 104.01317596435547,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.1099,
      "step": 900
    },
    {
      "epoch": 2.25,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.5458886027336121,
      "eval_runtime": 24.2511,
      "eval_samples_per_second": 2.886,
      "eval_steps_per_second": 2.886,
      "step": 900
    },
    {
      "epoch": 3.01,
      "grad_norm": 0.17340366542339325,
      "learning_rate": 1.3425925925925928e-05,
      "loss": 0.0997,
      "step": 910
    },
    {
      "epoch": 3.02,
      "grad_norm": 0.3285802900791168,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.0133,
      "step": 920
    },
    {
      "epoch": 3.02,
      "grad_norm": 0.04922209680080414,
      "learning_rate": 1.25e-05,
      "loss": 0.06,
      "step": 930
    },
    {
      "epoch": 3.03,
      "grad_norm": 0.21828210353851318,
      "learning_rate": 1.2037037037037037e-05,
      "loss": 0.0091,
      "step": 940
    },
    {
      "epoch": 3.04,
      "grad_norm": 2.8969061374664307,
      "learning_rate": 1.1574074074074075e-05,
      "loss": 0.0162,
      "step": 950
    },
    {
      "epoch": 3.05,
      "grad_norm": 0.06720645725727081,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0103,
      "step": 960
    },
    {
      "epoch": 3.06,
      "grad_norm": 0.17596295475959778,
      "learning_rate": 1.0648148148148148e-05,
      "loss": 0.017,
      "step": 970
    },
    {
      "epoch": 3.07,
      "grad_norm": 0.0713464543223381,
      "learning_rate": 1.0185185185185185e-05,
      "loss": 0.0262,
      "step": 980
    },
    {
      "epoch": 3.08,
      "grad_norm": 0.08627210557460785,
      "learning_rate": 9.722222222222223e-06,
      "loss": 0.0307,
      "step": 990
    },
    {
      "epoch": 3.08,
      "grad_norm": 0.07651841640472412,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.0101,
      "step": 1000
    },
    {
      "epoch": 3.09,
      "grad_norm": 2.683367967605591,
      "learning_rate": 8.796296296296297e-06,
      "loss": 0.0203,
      "step": 1010
    },
    {
      "epoch": 3.1,
      "grad_norm": 0.3468647301197052,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0139,
      "step": 1020
    },
    {
      "epoch": 3.11,
      "grad_norm": 0.22564736008644104,
      "learning_rate": 7.87037037037037e-06,
      "loss": 0.323,
      "step": 1030
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.05308224633336067,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.0047,
      "step": 1040
    },
    {
      "epoch": 3.12,
      "grad_norm": 0.052336595952510834,
      "learning_rate": 6.944444444444445e-06,
      "loss": 0.1145,
      "step": 1050
    },
    {
      "epoch": 3.13,
      "grad_norm": 161.04202270507812,
      "learning_rate": 6.481481481481481e-06,
      "loss": 0.3578,
      "step": 1060
    },
    {
      "epoch": 3.14,
      "grad_norm": 0.17146830260753632,
      "learning_rate": 6.0185185185185185e-06,
      "loss": 0.0197,
      "step": 1070
    },
    {
      "epoch": 3.15,
      "grad_norm": 0.23476547002792358,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0207,
      "step": 1080
    },
    {
      "epoch": 3.16,
      "grad_norm": 0.2218230664730072,
      "learning_rate": 5.092592592592592e-06,
      "loss": 0.5924,
      "step": 1090
    },
    {
      "epoch": 3.17,
      "grad_norm": 0.08028724044561386,
      "learning_rate": 4.6296296296296296e-06,
      "loss": 0.0047,
      "step": 1100
    },
    {
      "epoch": 3.17,
      "grad_norm": 0.07452563941478729,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.6085,
      "step": 1110
    },
    {
      "epoch": 3.18,
      "grad_norm": 0.06979314237833023,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.5321,
      "step": 1120
    },
    {
      "epoch": 3.19,
      "grad_norm": 0.08404794335365295,
      "learning_rate": 3.2407407407407406e-06,
      "loss": 0.0057,
      "step": 1130
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.08606185019016266,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.2438,
      "step": 1140
    },
    {
      "epoch": 3.21,
      "grad_norm": 0.06998346745967865,
      "learning_rate": 2.3148148148148148e-06,
      "loss": 0.0987,
      "step": 1150
    },
    {
      "epoch": 3.22,
      "grad_norm": 0.18059991300106049,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.3073,
      "step": 1160
    },
    {
      "epoch": 3.23,
      "grad_norm": 0.08045562356710434,
      "learning_rate": 1.388888888888889e-06,
      "loss": 0.2227,
      "step": 1170
    },
    {
      "epoch": 3.23,
      "grad_norm": 0.06636542081832886,
      "learning_rate": 9.259259259259259e-07,
      "loss": 0.7776,
      "step": 1180
    },
    {
      "epoch": 3.24,
      "grad_norm": 0.1821145862340927,
      "learning_rate": 4.6296296296296297e-07,
      "loss": 0.0074,
      "step": 1190
    },
    {
      "epoch": 3.25,
      "grad_norm": 0.10829867422580719,
      "learning_rate": 0.0,
      "loss": 1.0531,
      "step": 1200
    },
    {
      "epoch": 3.25,
      "eval_accuracy": 0.9285714285714286,
      "eval_loss": 0.2868471145629883,
      "eval_runtime": 17.7134,
      "eval_samples_per_second": 3.952,
      "eval_steps_per_second": 3.952,
      "step": 1200
    },
    {
      "epoch": 3.25,
      "step": 1200,
      "total_flos": 1.495384188125184e+18,
      "train_loss": 0.9583444615236173,
      "train_runtime": 830.113,
      "train_samples_per_second": 1.446,
      "train_steps_per_second": 1.446
    },
    {
      "epoch": 3.25,
      "eval_accuracy": 0.8516129032258064,
      "eval_loss": 0.5765717029571533,
      "eval_runtime": 43.5462,
      "eval_samples_per_second": 3.559,
      "eval_steps_per_second": 3.559,
      "step": 1200
    },
    {
      "epoch": 3.25,
      "eval_accuracy": 0.8516129032258064,
      "eval_loss": 0.5765716433525085,
      "eval_runtime": 41.7645,
      "eval_samples_per_second": 3.711,
      "eval_steps_per_second": 3.711,
      "step": 1200
    }
  ],
  "logging_steps": 10,
  "max_steps": 1200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 1.495384188125184e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}