{
  "best_metric": 0.9154929577464789,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-65-fold2/checkpoint-172",
  "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": 0.43661971830985913,
      "eval_loss": 0.7210970520973206,
      "eval_runtime": 0.9774,
      "eval_samples_per_second": 72.641,
      "eval_steps_per_second": 3.069,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.5070422535211268,
      "eval_loss": 0.7016032934188843,
      "eval_runtime": 0.9676,
      "eval_samples_per_second": 73.379,
      "eval_steps_per_second": 3.101,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.6197183098591549,
      "eval_loss": 0.6702623963356018,
      "eval_runtime": 0.9591,
      "eval_samples_per_second": 74.026,
      "eval_steps_per_second": 3.128,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 4.81842565536499,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.6946,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.6381291151046753,
      "eval_runtime": 0.9682,
      "eval_samples_per_second": 73.329,
      "eval_steps_per_second": 3.098,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.594517171382904,
      "eval_runtime": 0.9627,
      "eval_samples_per_second": 73.753,
      "eval_steps_per_second": 3.116,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.6083856225013733,
      "eval_runtime": 0.962,
      "eval_samples_per_second": 73.805,
      "eval_steps_per_second": 3.119,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 5.711062908172607,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6262,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5638889670372009,
      "eval_runtime": 0.9689,
      "eval_samples_per_second": 73.276,
      "eval_steps_per_second": 3.096,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5203454494476318,
      "eval_runtime": 0.9744,
      "eval_samples_per_second": 72.864,
      "eval_steps_per_second": 3.079,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4804993271827698,
      "eval_runtime": 0.9763,
      "eval_samples_per_second": 72.725,
      "eval_steps_per_second": 3.073,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 3.8290281295776367,
      "learning_rate": 5e-05,
      "loss": 0.544,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.5204476118087769,
      "eval_runtime": 0.9609,
      "eval_samples_per_second": 73.886,
      "eval_steps_per_second": 3.122,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.46348023414611816,
      "eval_runtime": 0.9654,
      "eval_samples_per_second": 73.545,
      "eval_steps_per_second": 3.108,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.4957320988178253,
      "eval_runtime": 0.9736,
      "eval_samples_per_second": 72.928,
      "eval_steps_per_second": 3.081,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 7.22072696685791,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.516,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.47228991985321045,
      "eval_runtime": 0.9665,
      "eval_samples_per_second": 73.461,
      "eval_steps_per_second": 3.104,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.5169680118560791,
      "eval_runtime": 0.9657,
      "eval_samples_per_second": 73.519,
      "eval_steps_per_second": 3.106,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5404752492904663,
      "eval_runtime": 0.9676,
      "eval_samples_per_second": 73.379,
      "eval_steps_per_second": 3.101,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 10.899679183959961,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4938,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.5082057118415833,
      "eval_runtime": 1.0054,
      "eval_samples_per_second": 70.619,
      "eval_steps_per_second": 2.984,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4608191251754761,
      "eval_runtime": 0.9655,
      "eval_samples_per_second": 73.539,
      "eval_steps_per_second": 3.107,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.4211063086986542,
      "eval_runtime": 0.9818,
      "eval_samples_per_second": 72.313,
      "eval_steps_per_second": 3.055,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 5.350540637969971,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.4123,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5014901757240295,
      "eval_runtime": 0.9712,
      "eval_samples_per_second": 73.105,
      "eval_steps_per_second": 3.089,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3935275375843048,
      "eval_runtime": 0.9719,
      "eval_samples_per_second": 73.054,
      "eval_steps_per_second": 3.087,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4178997874259949,
      "eval_runtime": 0.9729,
      "eval_samples_per_second": 72.978,
      "eval_steps_per_second": 3.084,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 4.967121124267578,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.3489,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.9014084507042254,
      "eval_loss": 0.3991130590438843,
      "eval_runtime": 0.9718,
      "eval_samples_per_second": 73.064,
      "eval_steps_per_second": 3.087,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.39102521538734436,
      "eval_runtime": 0.9741,
      "eval_samples_per_second": 72.89,
      "eval_steps_per_second": 3.08,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4276868402957916,
      "eval_runtime": 0.9687,
      "eval_samples_per_second": 73.295,
      "eval_steps_per_second": 3.097,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 6.507815837860107,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.2889,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.40316298604011536,
      "eval_runtime": 0.9728,
      "eval_samples_per_second": 72.986,
      "eval_steps_per_second": 3.084,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.37030014395713806,
      "eval_runtime": 0.9651,
      "eval_samples_per_second": 73.566,
      "eval_steps_per_second": 3.108,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.44038423895835876,
      "eval_runtime": 0.9745,
      "eval_samples_per_second": 72.856,
      "eval_steps_per_second": 3.078,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 4.317188739776611,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2659,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3666124641895294,
      "eval_runtime": 0.9751,
      "eval_samples_per_second": 72.814,
      "eval_steps_per_second": 3.077,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.39921075105667114,
      "eval_runtime": 0.9713,
      "eval_samples_per_second": 73.095,
      "eval_steps_per_second": 3.089,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4040003716945648,
      "eval_runtime": 0.9702,
      "eval_samples_per_second": 73.183,
      "eval_steps_per_second": 3.092,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 2.3390183448791504,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2269,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.355948269367218,
      "eval_runtime": 0.9741,
      "eval_samples_per_second": 72.89,
      "eval_steps_per_second": 3.08,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.4765550196170807,
      "eval_runtime": 0.9724,
      "eval_samples_per_second": 73.018,
      "eval_steps_per_second": 3.085,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3851951062679291,
      "eval_runtime": 0.965,
      "eval_samples_per_second": 73.578,
      "eval_steps_per_second": 3.109,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 4.255270957946777,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.2031,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.37017911672592163,
      "eval_runtime": 0.975,
      "eval_samples_per_second": 72.818,
      "eval_steps_per_second": 3.077,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.32030993700027466,
      "eval_runtime": 0.972,
      "eval_samples_per_second": 73.048,
      "eval_steps_per_second": 3.087,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5303316116333008,
      "eval_runtime": 0.9701,
      "eval_samples_per_second": 73.19,
      "eval_steps_per_second": 3.093,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 4.1614155769348145,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.2037,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3896830081939697,
      "eval_runtime": 0.9797,
      "eval_samples_per_second": 72.472,
      "eval_steps_per_second": 3.062,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3841053247451782,
      "eval_runtime": 0.9701,
      "eval_samples_per_second": 73.191,
      "eval_steps_per_second": 3.093,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.38963139057159424,
      "eval_runtime": 0.9753,
      "eval_samples_per_second": 72.802,
      "eval_steps_per_second": 3.076,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 3.8179051876068115,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.2018,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4176679849624634,
      "eval_runtime": 0.9825,
      "eval_samples_per_second": 72.267,
      "eval_steps_per_second": 3.054,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4547717273235321,
      "eval_runtime": 0.9725,
      "eval_samples_per_second": 73.008,
      "eval_steps_per_second": 3.085,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4115450382232666,
      "eval_runtime": 0.9703,
      "eval_samples_per_second": 73.173,
      "eval_steps_per_second": 3.092,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4120871424674988,
      "eval_runtime": 0.9779,
      "eval_samples_per_second": 72.604,
      "eval_steps_per_second": 3.068,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 6.913601875305176,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.1721,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3919922411441803,
      "eval_runtime": 0.9673,
      "eval_samples_per_second": 73.403,
      "eval_steps_per_second": 3.102,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.36926642060279846,
      "eval_runtime": 0.9653,
      "eval_samples_per_second": 73.552,
      "eval_steps_per_second": 3.108,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.36045223474502563,
      "eval_runtime": 0.957,
      "eval_samples_per_second": 74.188,
      "eval_steps_per_second": 3.135,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 4.530900001525879,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1678,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.54340660572052,
      "eval_runtime": 0.9636,
      "eval_samples_per_second": 73.681,
      "eval_steps_per_second": 3.113,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.41889968514442444,
      "eval_runtime": 0.9712,
      "eval_samples_per_second": 73.103,
      "eval_steps_per_second": 3.089,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.31237733364105225,
      "eval_runtime": 0.9625,
      "eval_samples_per_second": 73.765,
      "eval_steps_per_second": 3.117,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 3.286679744720459,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.1604,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.32932186126708984,
      "eval_runtime": 0.9716,
      "eval_samples_per_second": 73.073,
      "eval_steps_per_second": 3.088,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.9014084507042254,
      "eval_loss": 0.3371872901916504,
      "eval_runtime": 0.9747,
      "eval_samples_per_second": 72.843,
      "eval_steps_per_second": 3.078,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.35047727823257446,
      "eval_runtime": 0.9734,
      "eval_samples_per_second": 72.943,
      "eval_steps_per_second": 3.082,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 3.787102460861206,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.1406,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.9154929577464789,
      "eval_loss": 0.3094751834869385,
      "eval_runtime": 0.9675,
      "eval_samples_per_second": 73.386,
      "eval_steps_per_second": 3.101,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.9154929577464789,
      "eval_loss": 0.3054424822330475,
      "eval_runtime": 0.9688,
      "eval_samples_per_second": 73.284,
      "eval_steps_per_second": 3.097,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.36954671144485474,
      "eval_runtime": 0.9714,
      "eval_samples_per_second": 73.089,
      "eval_steps_per_second": 3.088,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 2.4381654262542725,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1492,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4058314263820648,
      "eval_runtime": 0.972,
      "eval_samples_per_second": 73.043,
      "eval_steps_per_second": 3.086,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4649871587753296,
      "eval_runtime": 0.9843,
      "eval_samples_per_second": 72.131,
      "eval_steps_per_second": 3.048,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4060044288635254,
      "eval_runtime": 0.9744,
      "eval_samples_per_second": 72.864,
      "eval_steps_per_second": 3.079,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 2.546844244003296,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1359,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.3819490671157837,
      "eval_runtime": 0.9781,
      "eval_samples_per_second": 72.586,
      "eval_steps_per_second": 3.067,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5229614973068237,
      "eval_runtime": 0.9718,
      "eval_samples_per_second": 73.063,
      "eval_steps_per_second": 3.087,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4986177980899811,
      "eval_runtime": 0.9746,
      "eval_samples_per_second": 72.849,
      "eval_steps_per_second": 3.078,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 2.6978113651275635,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1264,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4570479691028595,
      "eval_runtime": 0.9683,
      "eval_samples_per_second": 73.323,
      "eval_steps_per_second": 3.098,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.45072540640830994,
      "eval_runtime": 0.9768,
      "eval_samples_per_second": 72.685,
      "eval_steps_per_second": 3.071,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5765404105186462,
      "eval_runtime": 0.976,
      "eval_samples_per_second": 72.748,
      "eval_steps_per_second": 3.074,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 3.2018659114837646,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1478,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.45142102241516113,
      "eval_runtime": 0.9733,
      "eval_samples_per_second": 72.945,
      "eval_steps_per_second": 3.082,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.44336047768592834,
      "eval_runtime": 0.974,
      "eval_samples_per_second": 72.897,
      "eval_steps_per_second": 3.08,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.44032683968544006,
      "eval_runtime": 0.9712,
      "eval_samples_per_second": 73.106,
      "eval_steps_per_second": 3.089,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 3.408703327178955,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.1398,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5928447842597961,
      "eval_runtime": 0.9753,
      "eval_samples_per_second": 72.795,
      "eval_steps_per_second": 3.076,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.45865926146507263,
      "eval_runtime": 0.9755,
      "eval_samples_per_second": 72.781,
      "eval_steps_per_second": 3.075,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4053456783294678,
      "eval_runtime": 0.9772,
      "eval_samples_per_second": 72.658,
      "eval_steps_per_second": 3.07,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 2.2398223876953125,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.161,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4232625663280487,
      "eval_runtime": 0.9751,
      "eval_samples_per_second": 72.817,
      "eval_steps_per_second": 3.077,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4264310300350189,
      "eval_runtime": 0.9733,
      "eval_samples_per_second": 72.951,
      "eval_steps_per_second": 3.082,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4126971662044525,
      "eval_runtime": 0.9786,
      "eval_samples_per_second": 72.549,
      "eval_steps_per_second": 3.065,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 2.928635597229004,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1326,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.40130165219306946,
      "eval_runtime": 0.9713,
      "eval_samples_per_second": 73.095,
      "eval_steps_per_second": 3.089,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4389221966266632,
      "eval_runtime": 0.9794,
      "eval_samples_per_second": 72.493,
      "eval_steps_per_second": 3.063,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.3771601915359497,
      "eval_runtime": 1.0048,
      "eval_samples_per_second": 70.66,
      "eval_steps_per_second": 2.986,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 3.744046926498413,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1236,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3600108325481415,
      "eval_runtime": 0.9714,
      "eval_samples_per_second": 73.091,
      "eval_steps_per_second": 3.088,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.3890267312526703,
      "eval_runtime": 0.9697,
      "eval_samples_per_second": 73.219,
      "eval_steps_per_second": 3.094,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.4401465356349945,
      "eval_runtime": 0.9933,
      "eval_samples_per_second": 71.478,
      "eval_steps_per_second": 3.02,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 2.131753444671631,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.0973,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.40144455432891846,
      "eval_runtime": 0.9626,
      "eval_samples_per_second": 73.756,
      "eval_steps_per_second": 3.116,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.3766299784183502,
      "eval_runtime": 0.9827,
      "eval_samples_per_second": 72.248,
      "eval_steps_per_second": 3.053,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.3907908499240875,
      "eval_runtime": 0.9713,
      "eval_samples_per_second": 73.096,
      "eval_steps_per_second": 3.089,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.43390902876853943,
      "eval_runtime": 0.9782,
      "eval_samples_per_second": 72.584,
      "eval_steps_per_second": 3.067,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 1.7664235830307007,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1079,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.4567141532897949,
      "eval_runtime": 0.972,
      "eval_samples_per_second": 73.048,
      "eval_steps_per_second": 3.087,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.44145017862319946,
      "eval_runtime": 0.9751,
      "eval_samples_per_second": 72.813,
      "eval_steps_per_second": 3.077,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.41829437017440796,
      "eval_runtime": 0.9736,
      "eval_samples_per_second": 72.929,
      "eval_steps_per_second": 3.081,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 3.2624311447143555,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.1015,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.4039430618286133,
      "eval_runtime": 0.9659,
      "eval_samples_per_second": 73.504,
      "eval_steps_per_second": 3.106,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8873239436619719,
      "eval_loss": 0.3996119201183319,
      "eval_runtime": 0.9685,
      "eval_samples_per_second": 73.307,
      "eval_steps_per_second": 3.097,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.9014084507042254,
      "eval_loss": 0.403146356344223,
      "eval_runtime": 0.9729,
      "eval_samples_per_second": 72.978,
      "eval_steps_per_second": 3.084,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 3.2965073585510254,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.1174,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.41012877225875854,
      "eval_runtime": 0.9657,
      "eval_samples_per_second": 73.522,
      "eval_steps_per_second": 3.107,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.41530841588974,
      "eval_runtime": 0.9728,
      "eval_samples_per_second": 72.988,
      "eval_steps_per_second": 3.084,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.41457247734069824,
      "eval_runtime": 0.9792,
      "eval_samples_per_second": 72.506,
      "eval_steps_per_second": 3.064,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 1.616357445716858,
      "learning_rate": 0.0,
      "loss": 0.0968,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.4145110845565796,
      "eval_runtime": 0.9776,
      "eval_samples_per_second": 72.625,
      "eval_steps_per_second": 3.069,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 2.8402872494292173e+18,
      "train_loss": 0.24346253156661987,
      "train_runtime": 1610.5192,
      "train_samples_per_second": 24.65,
      "train_steps_per_second": 0.186
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.9154929577464789,
      "eval_loss": 0.3094751834869385,
      "eval_runtime": 1.0166,
      "eval_samples_per_second": 69.839,
      "eval_steps_per_second": 2.951,
      "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
}