{
  "best_metric": 0.9014084507042254,
  "best_model_checkpoint": "beit-base-patch16-224-65-fold2/checkpoint-100",
  "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": 0.5492957746478874,
      "eval_loss": 0.6829895973205566,
      "eval_runtime": 1.1518,
      "eval_samples_per_second": 61.641,
      "eval_steps_per_second": 2.605,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.647887323943662,
      "eval_loss": 0.6407253742218018,
      "eval_runtime": 1.1245,
      "eval_samples_per_second": 63.139,
      "eval_steps_per_second": 2.668,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.5352112676056338,
      "eval_loss": 0.6611723303794861,
      "eval_runtime": 1.0983,
      "eval_samples_per_second": 64.644,
      "eval_steps_per_second": 2.731,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 13.373871803283691,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.7094,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.6174536943435669,
      "eval_runtime": 1.0831,
      "eval_samples_per_second": 65.553,
      "eval_steps_per_second": 2.77,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.5914616584777832,
      "eval_runtime": 1.1133,
      "eval_samples_per_second": 63.775,
      "eval_steps_per_second": 2.695,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.5677279233932495,
      "eval_runtime": 1.1168,
      "eval_samples_per_second": 63.574,
      "eval_steps_per_second": 2.686,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 10.936724662780762,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6444,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5177403688430786,
      "eval_runtime": 1.0706,
      "eval_samples_per_second": 66.315,
      "eval_steps_per_second": 2.802,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4927891492843628,
      "eval_runtime": 1.0755,
      "eval_samples_per_second": 66.014,
      "eval_steps_per_second": 2.789,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.5641009211540222,
      "eval_runtime": 1.0927,
      "eval_samples_per_second": 64.977,
      "eval_steps_per_second": 2.745,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 10.08945083618164,
      "learning_rate": 5e-05,
      "loss": 0.5574,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.43716326355934143,
      "eval_runtime": 1.0928,
      "eval_samples_per_second": 64.973,
      "eval_steps_per_second": 2.745,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5677071809768677,
      "eval_runtime": 1.15,
      "eval_samples_per_second": 61.738,
      "eval_steps_per_second": 2.609,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5032159090042114,
      "eval_runtime": 1.1005,
      "eval_samples_per_second": 64.518,
      "eval_steps_per_second": 2.726,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 4.973866939544678,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.543,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.47447821497917175,
      "eval_runtime": 1.0767,
      "eval_samples_per_second": 65.942,
      "eval_steps_per_second": 2.786,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4056505262851715,
      "eval_runtime": 1.1102,
      "eval_samples_per_second": 63.953,
      "eval_steps_per_second": 2.702,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.4012921154499054,
      "eval_runtime": 1.0771,
      "eval_samples_per_second": 65.919,
      "eval_steps_per_second": 2.785,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 5.340551376342773,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4499,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.36697399616241455,
      "eval_runtime": 1.1391,
      "eval_samples_per_second": 62.328,
      "eval_steps_per_second": 2.634,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4215153157711029,
      "eval_runtime": 1.1192,
      "eval_samples_per_second": 63.435,
      "eval_steps_per_second": 2.68,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.4862484037876129,
      "eval_runtime": 1.0754,
      "eval_samples_per_second": 66.023,
      "eval_steps_per_second": 2.79,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 9.19057846069336,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3902,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.5781329274177551,
      "eval_runtime": 1.1009,
      "eval_samples_per_second": 64.491,
      "eval_steps_per_second": 2.725,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5248808860778809,
      "eval_runtime": 1.1057,
      "eval_samples_per_second": 64.211,
      "eval_steps_per_second": 2.713,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3936827778816223,
      "eval_runtime": 1.1333,
      "eval_samples_per_second": 62.651,
      "eval_steps_per_second": 2.647,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 3.5906007289886475,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.4029,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4131520092487335,
      "eval_runtime": 1.0935,
      "eval_samples_per_second": 64.93,
      "eval_steps_per_second": 2.744,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.41784021258354187,
      "eval_runtime": 1.1092,
      "eval_samples_per_second": 64.011,
      "eval_steps_per_second": 2.705,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.7273370027542114,
      "eval_runtime": 1.106,
      "eval_samples_per_second": 64.194,
      "eval_steps_per_second": 2.712,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 6.034952640533447,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.3163,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4221486747264862,
      "eval_runtime": 1.0943,
      "eval_samples_per_second": 64.883,
      "eval_steps_per_second": 2.742,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.4086194336414337,
      "eval_runtime": 1.1379,
      "eval_samples_per_second": 62.397,
      "eval_steps_per_second": 2.637,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.39464449882507324,
      "eval_runtime": 1.1089,
      "eval_samples_per_second": 64.025,
      "eval_steps_per_second": 2.705,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 6.2680439949035645,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2786,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5320041179656982,
      "eval_runtime": 1.1427,
      "eval_samples_per_second": 62.134,
      "eval_steps_per_second": 2.625,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.41318005323410034,
      "eval_runtime": 1.1147,
      "eval_samples_per_second": 63.695,
      "eval_steps_per_second": 2.691,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5542149543762207,
      "eval_runtime": 1.0924,
      "eval_samples_per_second": 64.994,
      "eval_steps_per_second": 2.746,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 5.675084114074707,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2763,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.9014084507042254,
      "eval_loss": 0.37337467074394226,
      "eval_runtime": 1.1285,
      "eval_samples_per_second": 62.913,
      "eval_steps_per_second": 2.658,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.44785401225090027,
      "eval_runtime": 1.0865,
      "eval_samples_per_second": 65.347,
      "eval_steps_per_second": 2.761,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3481605350971222,
      "eval_runtime": 1.0909,
      "eval_samples_per_second": 65.081,
      "eval_steps_per_second": 2.75,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 9.396614074707031,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.25,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5442134737968445,
      "eval_runtime": 1.0623,
      "eval_samples_per_second": 66.836,
      "eval_steps_per_second": 2.824,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.4211238920688629,
      "eval_runtime": 1.1024,
      "eval_samples_per_second": 64.406,
      "eval_steps_per_second": 2.721,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4860071539878845,
      "eval_runtime": 1.1377,
      "eval_samples_per_second": 62.405,
      "eval_steps_per_second": 2.637,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 3.626502513885498,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.2125,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.46536797285079956,
      "eval_runtime": 1.1169,
      "eval_samples_per_second": 63.568,
      "eval_steps_per_second": 2.686,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.477935254573822,
      "eval_runtime": 1.1116,
      "eval_samples_per_second": 63.871,
      "eval_steps_per_second": 2.699,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5691684484481812,
      "eval_runtime": 1.1081,
      "eval_samples_per_second": 64.071,
      "eval_steps_per_second": 2.707,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 6.315129280090332,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.2225,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4911651611328125,
      "eval_runtime": 1.097,
      "eval_samples_per_second": 64.719,
      "eval_steps_per_second": 2.735,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4528195559978485,
      "eval_runtime": 1.115,
      "eval_samples_per_second": 63.678,
      "eval_steps_per_second": 2.691,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4470233619213104,
      "eval_runtime": 1.1058,
      "eval_samples_per_second": 64.204,
      "eval_steps_per_second": 2.713,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.4251406788825989,
      "eval_runtime": 1.0717,
      "eval_samples_per_second": 66.249,
      "eval_steps_per_second": 2.799,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 6.046936988830566,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.1991,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.4864138662815094,
      "eval_runtime": 1.068,
      "eval_samples_per_second": 66.48,
      "eval_steps_per_second": 2.809,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4651569426059723,
      "eval_runtime": 1.0698,
      "eval_samples_per_second": 66.368,
      "eval_steps_per_second": 2.804,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5949256420135498,
      "eval_runtime": 1.1229,
      "eval_samples_per_second": 63.23,
      "eval_steps_per_second": 2.672,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 4.972904682159424,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.164,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 1.0035008192062378,
      "eval_runtime": 1.091,
      "eval_samples_per_second": 65.076,
      "eval_steps_per_second": 2.75,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6393434405326843,
      "eval_runtime": 1.1065,
      "eval_samples_per_second": 64.167,
      "eval_steps_per_second": 2.711,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.9221746921539307,
      "eval_runtime": 1.1189,
      "eval_samples_per_second": 63.454,
      "eval_steps_per_second": 2.681,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 31.394847869873047,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.1974,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 1.0632798671722412,
      "eval_runtime": 1.0728,
      "eval_samples_per_second": 66.183,
      "eval_steps_per_second": 2.796,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6050125360488892,
      "eval_runtime": 1.1003,
      "eval_samples_per_second": 64.529,
      "eval_steps_per_second": 2.727,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7133720517158508,
      "eval_runtime": 1.1005,
      "eval_samples_per_second": 64.514,
      "eval_steps_per_second": 2.726,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 4.075202465057373,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.213,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6648813486099243,
      "eval_runtime": 1.1101,
      "eval_samples_per_second": 63.956,
      "eval_steps_per_second": 2.702,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7125517129898071,
      "eval_runtime": 1.097,
      "eval_samples_per_second": 64.721,
      "eval_steps_per_second": 2.735,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.6906399130821228,
      "eval_runtime": 1.0988,
      "eval_samples_per_second": 64.617,
      "eval_steps_per_second": 2.73,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 4.072634220123291,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1642,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6955938339233398,
      "eval_runtime": 1.0988,
      "eval_samples_per_second": 64.617,
      "eval_steps_per_second": 2.73,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5828216075897217,
      "eval_runtime": 1.1089,
      "eval_samples_per_second": 64.03,
      "eval_steps_per_second": 2.705,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5865649580955505,
      "eval_runtime": 1.1049,
      "eval_samples_per_second": 64.258,
      "eval_steps_per_second": 2.715,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 9.487194061279297,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1657,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.6171615123748779,
      "eval_runtime": 1.1009,
      "eval_samples_per_second": 64.495,
      "eval_steps_per_second": 2.725,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7427834868431091,
      "eval_runtime": 1.0942,
      "eval_samples_per_second": 64.889,
      "eval_steps_per_second": 2.742,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.8981254696846008,
      "eval_runtime": 1.0972,
      "eval_samples_per_second": 64.708,
      "eval_steps_per_second": 2.734,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 6.037697792053223,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1347,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7168216705322266,
      "eval_runtime": 1.1259,
      "eval_samples_per_second": 63.063,
      "eval_steps_per_second": 2.665,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.8026114106178284,
      "eval_runtime": 1.113,
      "eval_samples_per_second": 63.789,
      "eval_steps_per_second": 2.695,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.8639427423477173,
      "eval_runtime": 1.1002,
      "eval_samples_per_second": 64.535,
      "eval_steps_per_second": 2.727,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 7.2083282470703125,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1335,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7604219913482666,
      "eval_runtime": 1.1013,
      "eval_samples_per_second": 64.472,
      "eval_steps_per_second": 2.724,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7992713451385498,
      "eval_runtime": 1.0706,
      "eval_samples_per_second": 66.321,
      "eval_steps_per_second": 2.802,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.833002507686615,
      "eval_runtime": 1.1411,
      "eval_samples_per_second": 62.218,
      "eval_steps_per_second": 2.629,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 4.454225540161133,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.145,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.8143167495727539,
      "eval_runtime": 1.0982,
      "eval_samples_per_second": 64.653,
      "eval_steps_per_second": 2.732,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.751991868019104,
      "eval_runtime": 1.1019,
      "eval_samples_per_second": 64.432,
      "eval_steps_per_second": 2.722,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.721619725227356,
      "eval_runtime": 1.0719,
      "eval_samples_per_second": 66.236,
      "eval_steps_per_second": 2.799,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 5.271942138671875,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.1658,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7968146204948425,
      "eval_runtime": 1.1157,
      "eval_samples_per_second": 63.638,
      "eval_steps_per_second": 2.689,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.7730194330215454,
      "eval_runtime": 1.1178,
      "eval_samples_per_second": 63.517,
      "eval_steps_per_second": 2.684,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.7449730634689331,
      "eval_runtime": 1.1284,
      "eval_samples_per_second": 62.921,
      "eval_steps_per_second": 2.659,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 4.669654846191406,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1381,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.7855215072631836,
      "eval_runtime": 1.0837,
      "eval_samples_per_second": 65.515,
      "eval_steps_per_second": 2.768,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.8253357410430908,
      "eval_runtime": 1.1299,
      "eval_samples_per_second": 62.836,
      "eval_steps_per_second": 2.655,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.8064681887626648,
      "eval_runtime": 1.0974,
      "eval_samples_per_second": 64.697,
      "eval_steps_per_second": 2.734,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 4.332691669464111,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1306,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7778077721595764,
      "eval_runtime": 1.1004,
      "eval_samples_per_second": 64.523,
      "eval_steps_per_second": 2.726,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7814451456069946,
      "eval_runtime": 1.1215,
      "eval_samples_per_second": 63.309,
      "eval_steps_per_second": 2.675,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.733501672744751,
      "eval_runtime": 1.1106,
      "eval_samples_per_second": 63.928,
      "eval_steps_per_second": 2.701,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 3.209725856781006,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.1027,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.7371585369110107,
      "eval_runtime": 1.0987,
      "eval_samples_per_second": 64.619,
      "eval_steps_per_second": 2.73,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.7617682814598083,
      "eval_runtime": 1.0988,
      "eval_samples_per_second": 64.618,
      "eval_steps_per_second": 2.73,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.7891401648521423,
      "eval_runtime": 1.0792,
      "eval_samples_per_second": 65.789,
      "eval_steps_per_second": 2.78,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.828731119632721,
      "eval_runtime": 1.1039,
      "eval_samples_per_second": 64.319,
      "eval_steps_per_second": 2.718,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 9.988916397094727,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1296,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.8412397503852844,
      "eval_runtime": 1.0672,
      "eval_samples_per_second": 66.527,
      "eval_steps_per_second": 2.811,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.8014044761657715,
      "eval_runtime": 1.1292,
      "eval_samples_per_second": 62.877,
      "eval_steps_per_second": 2.657,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.7529923915863037,
      "eval_runtime": 1.1018,
      "eval_samples_per_second": 64.442,
      "eval_steps_per_second": 2.723,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 5.536340236663818,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.1162,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.7242565155029297,
      "eval_runtime": 1.096,
      "eval_samples_per_second": 64.781,
      "eval_steps_per_second": 2.737,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.7246566414833069,
      "eval_runtime": 1.0962,
      "eval_samples_per_second": 64.771,
      "eval_steps_per_second": 2.737,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.735383927822113,
      "eval_runtime": 1.0904,
      "eval_samples_per_second": 65.113,
      "eval_steps_per_second": 2.751,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 3.339935064315796,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.1166,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7390521764755249,
      "eval_runtime": 1.0977,
      "eval_samples_per_second": 64.682,
      "eval_steps_per_second": 2.733,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7390065789222717,
      "eval_runtime": 1.0658,
      "eval_samples_per_second": 66.614,
      "eval_steps_per_second": 2.815,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7374176979064941,
      "eval_runtime": 1.0983,
      "eval_samples_per_second": 64.643,
      "eval_steps_per_second": 2.731,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 3.673130512237549,
      "learning_rate": 0.0,
      "loss": 0.1031,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7372613549232483,
      "eval_runtime": 1.1158,
      "eval_samples_per_second": 63.634,
      "eval_steps_per_second": 2.689,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 2.839022453308834e+18,
      "train_loss": 0.25909996310869854,
      "train_runtime": 1733.8719,
      "train_samples_per_second": 22.897,
      "train_steps_per_second": 0.173
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.9014084507042254,
      "eval_loss": 0.37337467074394226,
      "eval_runtime": 1.0981,
      "eval_samples_per_second": 64.659,
      "eval_steps_per_second": 2.732,
      "step": 300
    }
  ],
  "logging_steps": 10,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 2.839022453308834e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}