{
  "best_metric": 0.8732394366197183,
  "best_model_checkpoint": "beit-base-patch16-224-65-fold1/checkpoint-58",
  "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": 0.4647887323943662,
      "eval_loss": 0.7288300395011902,
      "eval_runtime": 1.1807,
      "eval_samples_per_second": 60.134,
      "eval_steps_per_second": 2.541,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.5633802816901409,
      "eval_loss": 0.6815673112869263,
      "eval_runtime": 1.0913,
      "eval_samples_per_second": 65.058,
      "eval_steps_per_second": 2.749,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.647887323943662,
      "eval_loss": 0.6401605010032654,
      "eval_runtime": 1.0922,
      "eval_samples_per_second": 65.008,
      "eval_steps_per_second": 2.747,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 5.065812587738037,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.699,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.6027737855911255,
      "eval_runtime": 1.0782,
      "eval_samples_per_second": 65.849,
      "eval_steps_per_second": 2.782,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.5877068042755127,
      "eval_runtime": 1.1218,
      "eval_samples_per_second": 63.289,
      "eval_steps_per_second": 2.674,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.5857630968093872,
      "eval_runtime": 1.1162,
      "eval_samples_per_second": 63.609,
      "eval_steps_per_second": 2.688,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 15.559165954589844,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.618,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.5098510980606079,
      "eval_runtime": 1.095,
      "eval_samples_per_second": 64.84,
      "eval_steps_per_second": 2.74,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.5926085710525513,
      "eval_runtime": 1.1178,
      "eval_samples_per_second": 63.519,
      "eval_steps_per_second": 2.684,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.48674505949020386,
      "eval_runtime": 1.0979,
      "eval_samples_per_second": 64.668,
      "eval_steps_per_second": 2.732,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 7.271378040313721,
      "learning_rate": 5e-05,
      "loss": 0.5564,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.473639577627182,
      "eval_runtime": 1.0987,
      "eval_samples_per_second": 64.619,
      "eval_steps_per_second": 2.73,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.4779972434043884,
      "eval_runtime": 1.1054,
      "eval_samples_per_second": 64.229,
      "eval_steps_per_second": 2.714,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.4812304675579071,
      "eval_runtime": 1.0871,
      "eval_samples_per_second": 65.311,
      "eval_steps_per_second": 2.76,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 5.897740364074707,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.5017,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4834084212779999,
      "eval_runtime": 1.1186,
      "eval_samples_per_second": 63.472,
      "eval_steps_per_second": 2.682,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4662351906299591,
      "eval_runtime": 1.1008,
      "eval_samples_per_second": 64.5,
      "eval_steps_per_second": 2.725,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.4560367465019226,
      "eval_runtime": 1.0958,
      "eval_samples_per_second": 64.795,
      "eval_steps_per_second": 2.738,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 7.2701735496521,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4603,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4046410918235779,
      "eval_runtime": 1.1011,
      "eval_samples_per_second": 64.48,
      "eval_steps_per_second": 2.725,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.33679458498954773,
      "eval_runtime": 1.0957,
      "eval_samples_per_second": 64.8,
      "eval_steps_per_second": 2.738,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3453676700592041,
      "eval_runtime": 1.104,
      "eval_samples_per_second": 64.312,
      "eval_steps_per_second": 2.717,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 9.926513671875,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.4037,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.6354864835739136,
      "eval_runtime": 1.0842,
      "eval_samples_per_second": 65.485,
      "eval_steps_per_second": 2.767,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.36235737800598145,
      "eval_runtime": 1.1051,
      "eval_samples_per_second": 64.245,
      "eval_steps_per_second": 2.715,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3748150169849396,
      "eval_runtime": 1.0996,
      "eval_samples_per_second": 64.57,
      "eval_steps_per_second": 2.728,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 5.17297887802124,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.3696,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.3798663020133972,
      "eval_runtime": 1.0807,
      "eval_samples_per_second": 65.699,
      "eval_steps_per_second": 2.776,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3885851800441742,
      "eval_runtime": 1.0949,
      "eval_samples_per_second": 64.848,
      "eval_steps_per_second": 2.74,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.34845802187919617,
      "eval_runtime": 1.093,
      "eval_samples_per_second": 64.961,
      "eval_steps_per_second": 2.745,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 7.104659557342529,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.3193,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.33851712942123413,
      "eval_runtime": 1.1028,
      "eval_samples_per_second": 64.379,
      "eval_steps_per_second": 2.72,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3690694272518158,
      "eval_runtime": 1.1165,
      "eval_samples_per_second": 63.589,
      "eval_steps_per_second": 2.687,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.38633036613464355,
      "eval_runtime": 1.1065,
      "eval_samples_per_second": 64.164,
      "eval_steps_per_second": 2.711,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 5.441641330718994,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.308,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3721933364868164,
      "eval_runtime": 1.0906,
      "eval_samples_per_second": 65.102,
      "eval_steps_per_second": 2.751,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3481193780899048,
      "eval_runtime": 1.0968,
      "eval_samples_per_second": 64.736,
      "eval_steps_per_second": 2.735,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3487893342971802,
      "eval_runtime": 1.1021,
      "eval_samples_per_second": 64.425,
      "eval_steps_per_second": 2.722,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 4.906981945037842,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2306,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.42525169253349304,
      "eval_runtime": 1.0966,
      "eval_samples_per_second": 64.747,
      "eval_steps_per_second": 2.736,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6243869662284851,
      "eval_runtime": 1.0613,
      "eval_samples_per_second": 66.897,
      "eval_steps_per_second": 2.827,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4688225984573364,
      "eval_runtime": 1.1002,
      "eval_samples_per_second": 64.533,
      "eval_steps_per_second": 2.727,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 3.6844112873077393,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.2431,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6079502701759338,
      "eval_runtime": 1.1011,
      "eval_samples_per_second": 64.483,
      "eval_steps_per_second": 2.725,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5795194506645203,
      "eval_runtime": 1.0742,
      "eval_samples_per_second": 66.097,
      "eval_steps_per_second": 2.793,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5478178858757019,
      "eval_runtime": 1.1175,
      "eval_samples_per_second": 63.536,
      "eval_steps_per_second": 2.685,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 12.373783111572266,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.2174,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5283036828041077,
      "eval_runtime": 1.0917,
      "eval_samples_per_second": 65.035,
      "eval_steps_per_second": 2.748,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5355969667434692,
      "eval_runtime": 1.0834,
      "eval_samples_per_second": 65.534,
      "eval_steps_per_second": 2.769,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5722697973251343,
      "eval_runtime": 1.1094,
      "eval_samples_per_second": 64.0,
      "eval_steps_per_second": 2.704,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 9.128671646118164,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.1928,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5488702058792114,
      "eval_runtime": 1.0861,
      "eval_samples_per_second": 65.369,
      "eval_steps_per_second": 2.762,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5277405977249146,
      "eval_runtime": 1.1056,
      "eval_samples_per_second": 64.219,
      "eval_steps_per_second": 2.713,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.4724764823913574,
      "eval_runtime": 1.1029,
      "eval_samples_per_second": 64.373,
      "eval_steps_per_second": 2.72,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.7954447865486145,
      "eval_runtime": 1.1388,
      "eval_samples_per_second": 62.345,
      "eval_steps_per_second": 2.634,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 12.15622615814209,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.1919,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5395719408988953,
      "eval_runtime": 1.1052,
      "eval_samples_per_second": 64.243,
      "eval_steps_per_second": 2.714,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6011538505554199,
      "eval_runtime": 1.1061,
      "eval_samples_per_second": 64.192,
      "eval_steps_per_second": 2.712,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.6191979050636292,
      "eval_runtime": 1.086,
      "eval_samples_per_second": 65.378,
      "eval_steps_per_second": 2.762,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 8.922574996948242,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1886,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.6233192682266235,
      "eval_runtime": 1.0874,
      "eval_samples_per_second": 65.294,
      "eval_steps_per_second": 2.759,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.6464739441871643,
      "eval_runtime": 1.0904,
      "eval_samples_per_second": 65.116,
      "eval_steps_per_second": 2.751,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7676438093185425,
      "eval_runtime": 1.1034,
      "eval_samples_per_second": 64.348,
      "eval_steps_per_second": 2.719,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 13.426934242248535,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.1661,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5266097187995911,
      "eval_runtime": 1.116,
      "eval_samples_per_second": 63.621,
      "eval_steps_per_second": 2.688,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.512731671333313,
      "eval_runtime": 1.1039,
      "eval_samples_per_second": 64.316,
      "eval_steps_per_second": 2.718,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5553781390190125,
      "eval_runtime": 1.1142,
      "eval_samples_per_second": 63.722,
      "eval_steps_per_second": 2.692,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 5.626668453216553,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.1746,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6332789063453674,
      "eval_runtime": 1.0946,
      "eval_samples_per_second": 64.864,
      "eval_steps_per_second": 2.741,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.606918454170227,
      "eval_runtime": 1.0944,
      "eval_samples_per_second": 64.877,
      "eval_steps_per_second": 2.741,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6963099241256714,
      "eval_runtime": 1.0772,
      "eval_samples_per_second": 65.909,
      "eval_steps_per_second": 2.785,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 3.974637031555176,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1487,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7241524457931519,
      "eval_runtime": 1.0986,
      "eval_samples_per_second": 64.63,
      "eval_steps_per_second": 2.731,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.8501065373420715,
      "eval_runtime": 1.0712,
      "eval_samples_per_second": 66.279,
      "eval_steps_per_second": 2.801,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6206911206245422,
      "eval_runtime": 1.0979,
      "eval_samples_per_second": 64.668,
      "eval_steps_per_second": 2.732,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 7.390707969665527,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1785,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5804906487464905,
      "eval_runtime": 1.078,
      "eval_samples_per_second": 65.865,
      "eval_steps_per_second": 2.783,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5997825264930725,
      "eval_runtime": 1.1123,
      "eval_samples_per_second": 63.829,
      "eval_steps_per_second": 2.697,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5253941416740417,
      "eval_runtime": 1.1215,
      "eval_samples_per_second": 63.307,
      "eval_steps_per_second": 2.675,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 5.779213905334473,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1711,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5723310708999634,
      "eval_runtime": 1.1096,
      "eval_samples_per_second": 63.988,
      "eval_steps_per_second": 2.704,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7298144698143005,
      "eval_runtime": 1.0892,
      "eval_samples_per_second": 65.186,
      "eval_steps_per_second": 2.754,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.664665699005127,
      "eval_runtime": 1.098,
      "eval_samples_per_second": 64.663,
      "eval_steps_per_second": 2.732,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 8.005589485168457,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1637,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7010626792907715,
      "eval_runtime": 1.1001,
      "eval_samples_per_second": 64.537,
      "eval_steps_per_second": 2.727,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7031036019325256,
      "eval_runtime": 1.1095,
      "eval_samples_per_second": 63.994,
      "eval_steps_per_second": 2.704,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7163373827934265,
      "eval_runtime": 1.0909,
      "eval_samples_per_second": 65.082,
      "eval_steps_per_second": 2.75,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 4.772933483123779,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.1618,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.6511116623878479,
      "eval_runtime": 1.0935,
      "eval_samples_per_second": 64.928,
      "eval_steps_per_second": 2.743,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6291322708129883,
      "eval_runtime": 1.0917,
      "eval_samples_per_second": 65.036,
      "eval_steps_per_second": 2.748,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.60442054271698,
      "eval_runtime": 1.123,
      "eval_samples_per_second": 63.226,
      "eval_steps_per_second": 2.672,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 3.7050230503082275,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.153,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5887604355812073,
      "eval_runtime": 1.1035,
      "eval_samples_per_second": 64.34,
      "eval_steps_per_second": 2.719,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5881189703941345,
      "eval_runtime": 1.0815,
      "eval_samples_per_second": 65.649,
      "eval_steps_per_second": 2.774,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.560383677482605,
      "eval_runtime": 1.0626,
      "eval_samples_per_second": 66.817,
      "eval_steps_per_second": 2.823,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 3.164400339126587,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1365,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.605511486530304,
      "eval_runtime": 1.0989,
      "eval_samples_per_second": 64.61,
      "eval_steps_per_second": 2.73,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.6326203942298889,
      "eval_runtime": 1.0959,
      "eval_samples_per_second": 64.785,
      "eval_steps_per_second": 2.737,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6685780882835388,
      "eval_runtime": 1.1107,
      "eval_samples_per_second": 63.921,
      "eval_steps_per_second": 2.701,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 2.557823419570923,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1231,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6954604983329773,
      "eval_runtime": 1.083,
      "eval_samples_per_second": 65.557,
      "eval_steps_per_second": 2.77,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7301970720291138,
      "eval_runtime": 1.0963,
      "eval_samples_per_second": 64.763,
      "eval_steps_per_second": 2.736,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7928118705749512,
      "eval_runtime": 1.0981,
      "eval_samples_per_second": 64.657,
      "eval_steps_per_second": 2.732,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 3.699218511581421,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.132,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7246793508529663,
      "eval_runtime": 1.0925,
      "eval_samples_per_second": 64.986,
      "eval_steps_per_second": 2.746,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7242569923400879,
      "eval_runtime": 1.1077,
      "eval_samples_per_second": 64.094,
      "eval_steps_per_second": 2.708,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7361048460006714,
      "eval_runtime": 1.0761,
      "eval_samples_per_second": 65.976,
      "eval_steps_per_second": 2.788,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7178685069084167,
      "eval_runtime": 1.0736,
      "eval_samples_per_second": 66.131,
      "eval_steps_per_second": 2.794,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 3.0329580307006836,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1126,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7054413557052612,
      "eval_runtime": 1.1062,
      "eval_samples_per_second": 64.186,
      "eval_steps_per_second": 2.712,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.719248354434967,
      "eval_runtime": 1.0785,
      "eval_samples_per_second": 65.834,
      "eval_steps_per_second": 2.782,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7357636094093323,
      "eval_runtime": 1.0957,
      "eval_samples_per_second": 64.799,
      "eval_steps_per_second": 2.738,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 5.440984725952148,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.1141,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7575234174728394,
      "eval_runtime": 1.0777,
      "eval_samples_per_second": 65.881,
      "eval_steps_per_second": 2.784,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7741107940673828,
      "eval_runtime": 1.101,
      "eval_samples_per_second": 64.486,
      "eval_steps_per_second": 2.725,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7877637147903442,
      "eval_runtime": 1.1237,
      "eval_samples_per_second": 63.183,
      "eval_steps_per_second": 2.67,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 4.872300624847412,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.1105,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7856650352478027,
      "eval_runtime": 1.0895,
      "eval_samples_per_second": 65.168,
      "eval_steps_per_second": 2.754,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7814451456069946,
      "eval_runtime": 1.1125,
      "eval_samples_per_second": 63.821,
      "eval_steps_per_second": 2.697,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.7784734964370728,
      "eval_runtime": 1.1178,
      "eval_samples_per_second": 63.516,
      "eval_steps_per_second": 2.684,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 3.307798385620117,
      "learning_rate": 0.0,
      "loss": 0.1006,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.778151273727417,
      "eval_runtime": 1.1114,
      "eval_samples_per_second": 63.881,
      "eval_steps_per_second": 2.699,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 2.839022453308834e+18,
      "train_loss": 0.25491045395533246,
      "train_runtime": 1727.8695,
      "train_samples_per_second": 22.976,
      "train_steps_per_second": 0.174
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3453676700592041,
      "eval_runtime": 1.0834,
      "eval_samples_per_second": 65.536,
      "eval_steps_per_second": 2.769,
      "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
}