{
  "best_metric": 0.8732394366197183,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-65-fold1/checkpoint-201",
  "epoch": 92.3076923076923,
  "eval_steps": 500,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9230769230769231,
      "eval_accuracy": 0.49295774647887325,
      "eval_loss": 0.7887781858444214,
      "eval_runtime": 1.0049,
      "eval_samples_per_second": 70.656,
      "eval_steps_per_second": 2.985,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.5070422535211268,
      "eval_loss": 0.7159498333930969,
      "eval_runtime": 0.9626,
      "eval_samples_per_second": 73.757,
      "eval_steps_per_second": 3.116,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.5070422535211268,
      "eval_loss": 0.7091380953788757,
      "eval_runtime": 0.9551,
      "eval_samples_per_second": 74.336,
      "eval_steps_per_second": 3.141,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 5.692840576171875,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.7703,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5352112676056338,
      "eval_loss": 0.6907549500465393,
      "eval_runtime": 0.9654,
      "eval_samples_per_second": 73.543,
      "eval_steps_per_second": 3.107,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.6197183098591549,
      "eval_loss": 0.6527045965194702,
      "eval_runtime": 0.978,
      "eval_samples_per_second": 72.596,
      "eval_steps_per_second": 3.067,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.6236115097999573,
      "eval_runtime": 0.972,
      "eval_samples_per_second": 73.044,
      "eval_steps_per_second": 3.086,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 4.818113327026367,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6435,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.6357130408287048,
      "eval_runtime": 0.9775,
      "eval_samples_per_second": 72.631,
      "eval_steps_per_second": 3.069,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.5442065596580505,
      "eval_runtime": 1.0045,
      "eval_samples_per_second": 70.683,
      "eval_steps_per_second": 2.987,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.5448713898658752,
      "eval_runtime": 0.9838,
      "eval_samples_per_second": 72.166,
      "eval_steps_per_second": 3.049,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 3.4911949634552,
      "learning_rate": 5e-05,
      "loss": 0.5366,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5123764872550964,
      "eval_runtime": 0.9995,
      "eval_samples_per_second": 71.034,
      "eval_steps_per_second": 3.001,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.5028589367866516,
      "eval_runtime": 0.9867,
      "eval_samples_per_second": 71.958,
      "eval_steps_per_second": 3.04,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.5486107468605042,
      "eval_runtime": 0.9813,
      "eval_samples_per_second": 72.355,
      "eval_steps_per_second": 3.057,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 6.1214070320129395,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.4577,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.5393596291542053,
      "eval_runtime": 0.9861,
      "eval_samples_per_second": 71.997,
      "eval_steps_per_second": 3.042,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5510964393615723,
      "eval_runtime": 0.9874,
      "eval_samples_per_second": 71.909,
      "eval_steps_per_second": 3.038,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.5794062614440918,
      "eval_runtime": 0.9882,
      "eval_samples_per_second": 71.849,
      "eval_steps_per_second": 3.036,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 4.4983015060424805,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4187,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.53675776720047,
      "eval_runtime": 1.0161,
      "eval_samples_per_second": 69.878,
      "eval_steps_per_second": 2.953,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4677673578262329,
      "eval_runtime": 0.9871,
      "eval_samples_per_second": 71.928,
      "eval_steps_per_second": 3.039,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.6597200036048889,
      "eval_runtime": 1.0221,
      "eval_samples_per_second": 69.468,
      "eval_steps_per_second": 2.935,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 4.643094539642334,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3542,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4969327747821808,
      "eval_runtime": 0.9908,
      "eval_samples_per_second": 71.661,
      "eval_steps_per_second": 3.028,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.7102980017662048,
      "eval_runtime": 0.9798,
      "eval_samples_per_second": 72.463,
      "eval_steps_per_second": 3.062,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.49788960814476013,
      "eval_runtime": 0.9857,
      "eval_samples_per_second": 72.032,
      "eval_steps_per_second": 3.044,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 3.1563220024108887,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.3057,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.5270604491233826,
      "eval_runtime": 0.9884,
      "eval_samples_per_second": 71.835,
      "eval_steps_per_second": 3.035,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5357353091239929,
      "eval_runtime": 0.9814,
      "eval_samples_per_second": 72.342,
      "eval_steps_per_second": 3.057,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.48468706011772156,
      "eval_runtime": 0.9993,
      "eval_samples_per_second": 71.047,
      "eval_steps_per_second": 3.002,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 6.270655632019043,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.2816,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5424976944923401,
      "eval_runtime": 0.9807,
      "eval_samples_per_second": 72.4,
      "eval_steps_per_second": 3.059,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5239164233207703,
      "eval_runtime": 1.0158,
      "eval_samples_per_second": 69.897,
      "eval_steps_per_second": 2.953,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4140526354312897,
      "eval_runtime": 0.9875,
      "eval_samples_per_second": 71.897,
      "eval_steps_per_second": 3.038,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 3.314903497695923,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2881,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.4996558725833893,
      "eval_runtime": 0.9975,
      "eval_samples_per_second": 71.178,
      "eval_steps_per_second": 3.008,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.42159879207611084,
      "eval_runtime": 0.983,
      "eval_samples_per_second": 72.227,
      "eval_steps_per_second": 3.052,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.46680712699890137,
      "eval_runtime": 0.9908,
      "eval_samples_per_second": 71.663,
      "eval_steps_per_second": 3.028,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 2.419731616973877,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2421,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5904483199119568,
      "eval_runtime": 0.9786,
      "eval_samples_per_second": 72.553,
      "eval_steps_per_second": 3.066,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5239621996879578,
      "eval_runtime": 0.9902,
      "eval_samples_per_second": 71.7,
      "eval_steps_per_second": 3.03,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.99370938539505,
      "eval_runtime": 0.9918,
      "eval_samples_per_second": 71.586,
      "eval_steps_per_second": 3.025,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 2.3661601543426514,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.2402,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.4988890588283539,
      "eval_runtime": 0.9897,
      "eval_samples_per_second": 71.739,
      "eval_steps_per_second": 3.031,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7231740951538086,
      "eval_runtime": 0.9872,
      "eval_samples_per_second": 71.923,
      "eval_steps_per_second": 3.039,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4815123677253723,
      "eval_runtime": 0.9925,
      "eval_samples_per_second": 71.533,
      "eval_steps_per_second": 3.023,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 2.6437292098999023,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1862,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7431491017341614,
      "eval_runtime": 0.9913,
      "eval_samples_per_second": 71.622,
      "eval_steps_per_second": 3.026,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.44340020418167114,
      "eval_runtime": 0.9934,
      "eval_samples_per_second": 71.475,
      "eval_steps_per_second": 3.02,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4759599268436432,
      "eval_runtime": 0.9853,
      "eval_samples_per_second": 72.061,
      "eval_steps_per_second": 3.045,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 2.1550378799438477,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.1783,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5006221532821655,
      "eval_runtime": 0.9897,
      "eval_samples_per_second": 71.737,
      "eval_steps_per_second": 3.031,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.49856680631637573,
      "eval_runtime": 0.9788,
      "eval_samples_per_second": 72.535,
      "eval_steps_per_second": 3.065,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.794731616973877,
      "eval_runtime": 0.9805,
      "eval_samples_per_second": 72.414,
      "eval_steps_per_second": 3.06,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.48966798186302185,
      "eval_runtime": 0.9856,
      "eval_samples_per_second": 72.035,
      "eval_steps_per_second": 3.044,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 4.09879207611084,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.1685,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.7500377893447876,
      "eval_runtime": 0.9859,
      "eval_samples_per_second": 72.013,
      "eval_steps_per_second": 3.043,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6052554249763489,
      "eval_runtime": 0.9867,
      "eval_samples_per_second": 71.954,
      "eval_steps_per_second": 3.04,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4776698350906372,
      "eval_runtime": 0.9829,
      "eval_samples_per_second": 72.233,
      "eval_steps_per_second": 3.052,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 2.7736315727233887,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1779,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5799605250358582,
      "eval_runtime": 0.9783,
      "eval_samples_per_second": 72.578,
      "eval_steps_per_second": 3.067,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4681435227394104,
      "eval_runtime": 0.984,
      "eval_samples_per_second": 72.155,
      "eval_steps_per_second": 3.049,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7729381322860718,
      "eval_runtime": 0.9844,
      "eval_samples_per_second": 72.123,
      "eval_steps_per_second": 3.047,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 6.783754825592041,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.1502,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.6486974358558655,
      "eval_runtime": 0.9835,
      "eval_samples_per_second": 72.191,
      "eval_steps_per_second": 3.05,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5224327445030212,
      "eval_runtime": 0.982,
      "eval_samples_per_second": 72.3,
      "eval_steps_per_second": 3.055,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7017256021499634,
      "eval_runtime": 1.0037,
      "eval_samples_per_second": 70.739,
      "eval_steps_per_second": 2.989,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 2.59287428855896,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.1586,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.603446364402771,
      "eval_runtime": 1.0504,
      "eval_samples_per_second": 67.596,
      "eval_steps_per_second": 2.856,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5791419744491577,
      "eval_runtime": 1.0306,
      "eval_samples_per_second": 68.891,
      "eval_steps_per_second": 2.911,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.565058708190918,
      "eval_runtime": 0.9801,
      "eval_samples_per_second": 72.445,
      "eval_steps_per_second": 3.061,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 3.497166872024536,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.134,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.486237496137619,
      "eval_runtime": 0.98,
      "eval_samples_per_second": 72.446,
      "eval_steps_per_second": 3.061,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.6750636100769043,
      "eval_runtime": 0.9858,
      "eval_samples_per_second": 72.026,
      "eval_steps_per_second": 3.043,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5924672484397888,
      "eval_runtime": 1.0279,
      "eval_samples_per_second": 69.075,
      "eval_steps_per_second": 2.919,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 1.8251292705535889,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1602,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.39819350838661194,
      "eval_runtime": 0.9991,
      "eval_samples_per_second": 71.061,
      "eval_steps_per_second": 3.003,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5968966484069824,
      "eval_runtime": 1.0038,
      "eval_samples_per_second": 70.73,
      "eval_steps_per_second": 2.989,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5720605254173279,
      "eval_runtime": 0.9693,
      "eval_samples_per_second": 73.252,
      "eval_steps_per_second": 3.095,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 1.8045357465744019,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1217,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.38157418370246887,
      "eval_runtime": 0.9841,
      "eval_samples_per_second": 72.148,
      "eval_steps_per_second": 3.048,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4109618365764618,
      "eval_runtime": 0.9741,
      "eval_samples_per_second": 72.89,
      "eval_steps_per_second": 3.08,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6715837717056274,
      "eval_runtime": 0.9934,
      "eval_samples_per_second": 71.469,
      "eval_steps_per_second": 3.02,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 4.87626838684082,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1274,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.34985408186912537,
      "eval_runtime": 0.9766,
      "eval_samples_per_second": 72.704,
      "eval_steps_per_second": 3.072,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.36712145805358887,
      "eval_runtime": 0.9825,
      "eval_samples_per_second": 72.266,
      "eval_steps_per_second": 3.053,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5317980051040649,
      "eval_runtime": 0.9694,
      "eval_samples_per_second": 73.245,
      "eval_steps_per_second": 3.095,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 1.64058518409729,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.1277,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6733711361885071,
      "eval_runtime": 0.9874,
      "eval_samples_per_second": 71.905,
      "eval_steps_per_second": 3.038,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.47263458371162415,
      "eval_runtime": 0.9771,
      "eval_samples_per_second": 72.665,
      "eval_steps_per_second": 3.07,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4311037063598633,
      "eval_runtime": 0.9776,
      "eval_samples_per_second": 72.63,
      "eval_steps_per_second": 3.069,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 2.334254026412964,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.1232,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7072361707687378,
      "eval_runtime": 0.969,
      "eval_samples_per_second": 73.27,
      "eval_steps_per_second": 3.096,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5859442949295044,
      "eval_runtime": 0.9757,
      "eval_samples_per_second": 72.767,
      "eval_steps_per_second": 3.075,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.3757818341255188,
      "eval_runtime": 0.9776,
      "eval_samples_per_second": 72.628,
      "eval_steps_per_second": 3.069,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 1.5687333345413208,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1293,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.3672768175601959,
      "eval_runtime": 0.977,
      "eval_samples_per_second": 72.668,
      "eval_steps_per_second": 3.07,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.367299884557724,
      "eval_runtime": 0.9773,
      "eval_samples_per_second": 72.646,
      "eval_steps_per_second": 3.07,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.47515830397605896,
      "eval_runtime": 0.9705,
      "eval_samples_per_second": 73.158,
      "eval_steps_per_second": 3.091,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 2.638277053833008,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1117,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.44498294591903687,
      "eval_runtime": 0.9801,
      "eval_samples_per_second": 72.44,
      "eval_steps_per_second": 3.061,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.44371137022972107,
      "eval_runtime": 0.9778,
      "eval_samples_per_second": 72.608,
      "eval_steps_per_second": 3.068,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4329843521118164,
      "eval_runtime": 0.9979,
      "eval_samples_per_second": 71.146,
      "eval_steps_per_second": 3.006,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 2.258843421936035,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.1092,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5095254182815552,
      "eval_runtime": 0.9679,
      "eval_samples_per_second": 73.356,
      "eval_steps_per_second": 3.1,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.49475494027137756,
      "eval_runtime": 0.9923,
      "eval_samples_per_second": 71.549,
      "eval_steps_per_second": 3.023,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.413480281829834,
      "eval_runtime": 0.9737,
      "eval_samples_per_second": 72.914,
      "eval_steps_per_second": 3.081,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4189954698085785,
      "eval_runtime": 0.9819,
      "eval_samples_per_second": 72.311,
      "eval_steps_per_second": 3.055,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 2.083444595336914,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1151,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.41941961646080017,
      "eval_runtime": 0.9662,
      "eval_samples_per_second": 73.482,
      "eval_steps_per_second": 3.105,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.43557149171829224,
      "eval_runtime": 0.9875,
      "eval_samples_per_second": 71.901,
      "eval_steps_per_second": 3.038,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.46227431297302246,
      "eval_runtime": 0.9652,
      "eval_samples_per_second": 73.558,
      "eval_steps_per_second": 3.108,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 3.552259683609009,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.1085,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.48452523350715637,
      "eval_runtime": 0.9766,
      "eval_samples_per_second": 72.702,
      "eval_steps_per_second": 3.072,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.49984505772590637,
      "eval_runtime": 0.9733,
      "eval_samples_per_second": 72.949,
      "eval_steps_per_second": 3.082,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5181298851966858,
      "eval_runtime": 0.9778,
      "eval_samples_per_second": 72.612,
      "eval_steps_per_second": 3.068,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 3.29520583152771,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.0908,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5373069643974304,
      "eval_runtime": 0.9643,
      "eval_samples_per_second": 73.627,
      "eval_steps_per_second": 3.111,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5465375185012817,
      "eval_runtime": 0.9916,
      "eval_samples_per_second": 71.6,
      "eval_steps_per_second": 3.025,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5421547889709473,
      "eval_runtime": 0.9804,
      "eval_samples_per_second": 72.422,
      "eval_steps_per_second": 3.06,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 1.8515039682388306,
      "learning_rate": 0.0,
      "loss": 0.0902,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5417219996452332,
      "eval_runtime": 0.9779,
      "eval_samples_per_second": 72.602,
      "eval_steps_per_second": 3.068,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 2.8402872494292173e+18,
      "train_loss": 0.23691350758075713,
      "train_runtime": 1617.7362,
      "train_samples_per_second": 24.54,
      "train_steps_per_second": 0.185
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.38157418370246887,
      "eval_runtime": 1.0366,
      "eval_samples_per_second": 68.495,
      "eval_steps_per_second": 2.894,
      "step": 300
    }
  ],
  "logging_steps": 10,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.8402872494292173e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}