{
  "best_metric": 0.8732394366197183,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-65-fold4/checkpoint-188",
  "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": 0.7218557596206665,
      "eval_runtime": 0.9745,
      "eval_samples_per_second": 72.855,
      "eval_steps_per_second": 3.078,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.5492957746478874,
      "eval_loss": 0.686266303062439,
      "eval_runtime": 0.9669,
      "eval_samples_per_second": 73.432,
      "eval_steps_per_second": 3.103,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.5492957746478874,
      "eval_loss": 0.6578378081321716,
      "eval_runtime": 0.9657,
      "eval_samples_per_second": 73.525,
      "eval_steps_per_second": 3.107,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 3.424938678741455,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.6825,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6338028169014085,
      "eval_loss": 0.6292961239814758,
      "eval_runtime": 0.9785,
      "eval_samples_per_second": 72.558,
      "eval_steps_per_second": 3.066,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.6186050176620483,
      "eval_runtime": 0.987,
      "eval_samples_per_second": 71.936,
      "eval_steps_per_second": 3.04,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.6135374903678894,
      "eval_runtime": 1.0113,
      "eval_samples_per_second": 70.204,
      "eval_steps_per_second": 2.966,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 4.245552062988281,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6206,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.647887323943662,
      "eval_loss": 0.6162843704223633,
      "eval_runtime": 0.9805,
      "eval_samples_per_second": 72.413,
      "eval_steps_per_second": 3.06,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.647887323943662,
      "eval_loss": 0.6350468397140503,
      "eval_runtime": 0.9932,
      "eval_samples_per_second": 71.487,
      "eval_steps_per_second": 3.021,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.6078002452850342,
      "eval_runtime": 0.9899,
      "eval_samples_per_second": 71.723,
      "eval_steps_per_second": 3.031,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 3.0329267978668213,
      "learning_rate": 5e-05,
      "loss": 0.5728,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.6872609257698059,
      "eval_runtime": 0.9874,
      "eval_samples_per_second": 71.907,
      "eval_steps_per_second": 3.038,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.6771360635757446,
      "eval_runtime": 0.984,
      "eval_samples_per_second": 72.152,
      "eval_steps_per_second": 3.049,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.5912012457847595,
      "eval_runtime": 0.9919,
      "eval_samples_per_second": 71.581,
      "eval_steps_per_second": 3.025,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 5.794826030731201,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.5329,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5523556470870972,
      "eval_runtime": 0.9889,
      "eval_samples_per_second": 71.799,
      "eval_steps_per_second": 3.034,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.5922580361366272,
      "eval_runtime": 0.9948,
      "eval_samples_per_second": 71.37,
      "eval_steps_per_second": 3.016,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.6649972200393677,
      "eval_runtime": 0.9954,
      "eval_samples_per_second": 71.326,
      "eval_steps_per_second": 3.014,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 4.871856689453125,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4279,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5182512402534485,
      "eval_runtime": 0.9953,
      "eval_samples_per_second": 71.338,
      "eval_steps_per_second": 3.014,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4760647714138031,
      "eval_runtime": 0.9935,
      "eval_samples_per_second": 71.467,
      "eval_steps_per_second": 3.02,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.558963418006897,
      "eval_runtime": 0.9911,
      "eval_samples_per_second": 71.637,
      "eval_steps_per_second": 3.027,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 4.168473243713379,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.4055,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5320470333099365,
      "eval_runtime": 1.0156,
      "eval_samples_per_second": 69.91,
      "eval_steps_per_second": 2.954,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.6604605913162231,
      "eval_runtime": 0.9924,
      "eval_samples_per_second": 71.546,
      "eval_steps_per_second": 3.023,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5820997357368469,
      "eval_runtime": 0.9824,
      "eval_samples_per_second": 72.271,
      "eval_steps_per_second": 3.054,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 5.980050086975098,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.3478,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5588834881782532,
      "eval_runtime": 0.9895,
      "eval_samples_per_second": 71.75,
      "eval_steps_per_second": 3.032,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.6247106194496155,
      "eval_runtime": 0.9919,
      "eval_samples_per_second": 71.578,
      "eval_steps_per_second": 3.024,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.7006160020828247,
      "eval_runtime": 1.0036,
      "eval_samples_per_second": 70.747,
      "eval_steps_per_second": 2.989,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 3.0665063858032227,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.3769,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.7575426697731018,
      "eval_runtime": 0.9915,
      "eval_samples_per_second": 71.611,
      "eval_steps_per_second": 3.026,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5367358922958374,
      "eval_runtime": 0.991,
      "eval_samples_per_second": 71.646,
      "eval_steps_per_second": 3.027,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.533458948135376,
      "eval_runtime": 0.9946,
      "eval_samples_per_second": 71.388,
      "eval_steps_per_second": 3.016,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 5.900879859924316,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2957,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5913785696029663,
      "eval_runtime": 1.0059,
      "eval_samples_per_second": 70.582,
      "eval_steps_per_second": 2.982,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.6780489683151245,
      "eval_runtime": 0.9943,
      "eval_samples_per_second": 71.404,
      "eval_steps_per_second": 3.017,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.534493625164032,
      "eval_runtime": 0.9982,
      "eval_samples_per_second": 71.129,
      "eval_steps_per_second": 3.005,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 3.5530407428741455,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2463,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.613211989402771,
      "eval_runtime": 0.9924,
      "eval_samples_per_second": 71.54,
      "eval_steps_per_second": 3.023,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5757858157157898,
      "eval_runtime": 0.9882,
      "eval_samples_per_second": 71.845,
      "eval_steps_per_second": 3.036,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.7235603332519531,
      "eval_runtime": 0.9977,
      "eval_samples_per_second": 71.165,
      "eval_steps_per_second": 3.007,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 4.502363204956055,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.2323,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5247332453727722,
      "eval_runtime": 1.0185,
      "eval_samples_per_second": 69.714,
      "eval_steps_per_second": 2.946,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.6018014550209045,
      "eval_runtime": 0.994,
      "eval_samples_per_second": 71.429,
      "eval_steps_per_second": 3.018,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5366302728652954,
      "eval_runtime": 0.9895,
      "eval_samples_per_second": 71.751,
      "eval_steps_per_second": 3.032,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 1.8001881837844849,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1921,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.6314247846603394,
      "eval_runtime": 0.9859,
      "eval_samples_per_second": 72.013,
      "eval_steps_per_second": 3.043,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5763274431228638,
      "eval_runtime": 0.9986,
      "eval_samples_per_second": 71.096,
      "eval_steps_per_second": 3.004,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5573564171791077,
      "eval_runtime": 0.9999,
      "eval_samples_per_second": 71.009,
      "eval_steps_per_second": 3.0,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 2.9866695404052734,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.1686,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6260572075843811,
      "eval_runtime": 1.0023,
      "eval_samples_per_second": 70.835,
      "eval_steps_per_second": 2.993,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6524580121040344,
      "eval_runtime": 0.9907,
      "eval_samples_per_second": 71.669,
      "eval_steps_per_second": 3.028,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5726364254951477,
      "eval_runtime": 0.9944,
      "eval_samples_per_second": 71.399,
      "eval_steps_per_second": 3.017,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.8199837803840637,
      "eval_runtime": 0.9956,
      "eval_samples_per_second": 71.317,
      "eval_steps_per_second": 3.013,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 8.961894035339355,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.2073,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4798416793346405,
      "eval_runtime": 0.9976,
      "eval_samples_per_second": 71.172,
      "eval_steps_per_second": 3.007,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5341726541519165,
      "eval_runtime": 0.9976,
      "eval_samples_per_second": 71.169,
      "eval_steps_per_second": 3.007,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.48342809081077576,
      "eval_runtime": 0.9955,
      "eval_samples_per_second": 71.324,
      "eval_steps_per_second": 3.014,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 6.036099433898926,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1702,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.6100845336914062,
      "eval_runtime": 1.0014,
      "eval_samples_per_second": 70.903,
      "eval_steps_per_second": 2.996,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.47786733508110046,
      "eval_runtime": 0.9939,
      "eval_samples_per_second": 71.433,
      "eval_steps_per_second": 3.018,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5047819018363953,
      "eval_runtime": 0.9956,
      "eval_samples_per_second": 71.313,
      "eval_steps_per_second": 3.013,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 3.7059566974639893,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.153,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.6297626495361328,
      "eval_runtime": 0.9916,
      "eval_samples_per_second": 71.6,
      "eval_steps_per_second": 3.025,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5995147824287415,
      "eval_runtime": 0.9924,
      "eval_samples_per_second": 71.541,
      "eval_steps_per_second": 3.023,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.6475109457969666,
      "eval_runtime": 0.9899,
      "eval_samples_per_second": 71.724,
      "eval_steps_per_second": 3.031,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 2.797576427459717,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.1508,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.48875874280929565,
      "eval_runtime": 1.0078,
      "eval_samples_per_second": 70.449,
      "eval_steps_per_second": 2.977,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4953955411911011,
      "eval_runtime": 0.9916,
      "eval_samples_per_second": 71.604,
      "eval_steps_per_second": 3.026,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.43896353244781494,
      "eval_runtime": 0.9963,
      "eval_samples_per_second": 71.26,
      "eval_steps_per_second": 3.011,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 2.899003744125366,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1293,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.477837473154068,
      "eval_runtime": 0.9883,
      "eval_samples_per_second": 71.838,
      "eval_steps_per_second": 3.035,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4887799024581909,
      "eval_runtime": 0.9913,
      "eval_samples_per_second": 71.622,
      "eval_steps_per_second": 3.026,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.48321232199668884,
      "eval_runtime": 0.9931,
      "eval_samples_per_second": 71.494,
      "eval_steps_per_second": 3.021,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 2.5362701416015625,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1489,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.527686595916748,
      "eval_runtime": 0.981,
      "eval_samples_per_second": 72.378,
      "eval_steps_per_second": 3.058,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.6216662526130676,
      "eval_runtime": 0.9828,
      "eval_samples_per_second": 72.243,
      "eval_steps_per_second": 3.053,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.6089593768119812,
      "eval_runtime": 0.9925,
      "eval_samples_per_second": 71.539,
      "eval_steps_per_second": 3.023,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 2.5179524421691895,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1487,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.5424482226371765,
      "eval_runtime": 0.9877,
      "eval_samples_per_second": 71.886,
      "eval_steps_per_second": 3.037,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.5569549202919006,
      "eval_runtime": 0.9873,
      "eval_samples_per_second": 71.913,
      "eval_steps_per_second": 3.039,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.7247959971427917,
      "eval_runtime": 0.9969,
      "eval_samples_per_second": 71.22,
      "eval_steps_per_second": 3.009,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 3.1306471824645996,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1456,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5840876698493958,
      "eval_runtime": 1.0357,
      "eval_samples_per_second": 68.554,
      "eval_steps_per_second": 2.897,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5905009508132935,
      "eval_runtime": 1.1058,
      "eval_samples_per_second": 64.208,
      "eval_steps_per_second": 2.713,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5608553290367126,
      "eval_runtime": 1.0367,
      "eval_samples_per_second": 68.484,
      "eval_steps_per_second": 2.894,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 2.4782862663269043,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.1284,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.546994149684906,
      "eval_runtime": 0.9751,
      "eval_samples_per_second": 72.81,
      "eval_steps_per_second": 3.076,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5472754240036011,
      "eval_runtime": 1.0047,
      "eval_samples_per_second": 70.665,
      "eval_steps_per_second": 2.986,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5813232660293579,
      "eval_runtime": 0.9977,
      "eval_samples_per_second": 71.162,
      "eval_steps_per_second": 3.007,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 1.8887909650802612,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.1225,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.5682988166809082,
      "eval_runtime": 0.9949,
      "eval_samples_per_second": 71.367,
      "eval_steps_per_second": 3.016,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5581419467926025,
      "eval_runtime": 0.9914,
      "eval_samples_per_second": 71.615,
      "eval_steps_per_second": 3.026,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5717195272445679,
      "eval_runtime": 0.976,
      "eval_samples_per_second": 72.745,
      "eval_steps_per_second": 3.074,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 2.5398831367492676,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1233,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.6054385304450989,
      "eval_runtime": 0.9717,
      "eval_samples_per_second": 73.067,
      "eval_steps_per_second": 3.087,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5910421013832092,
      "eval_runtime": 0.9862,
      "eval_samples_per_second": 71.995,
      "eval_steps_per_second": 3.042,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.570685863494873,
      "eval_runtime": 0.9929,
      "eval_samples_per_second": 71.507,
      "eval_steps_per_second": 3.021,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 2.284491539001465,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1234,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5732807517051697,
      "eval_runtime": 0.9877,
      "eval_samples_per_second": 71.884,
      "eval_steps_per_second": 3.037,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5748334527015686,
      "eval_runtime": 0.9897,
      "eval_samples_per_second": 71.74,
      "eval_steps_per_second": 3.031,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5722648501396179,
      "eval_runtime": 0.9949,
      "eval_samples_per_second": 71.363,
      "eval_steps_per_second": 3.015,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 4.123226165771484,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.1219,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5503106713294983,
      "eval_runtime": 1.0099,
      "eval_samples_per_second": 70.307,
      "eval_steps_per_second": 2.971,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5531720519065857,
      "eval_runtime": 0.978,
      "eval_samples_per_second": 72.6,
      "eval_steps_per_second": 3.068,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5827966332435608,
      "eval_runtime": 0.9983,
      "eval_samples_per_second": 71.119,
      "eval_steps_per_second": 3.005,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6061961650848389,
      "eval_runtime": 0.98,
      "eval_samples_per_second": 72.449,
      "eval_steps_per_second": 3.061,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 2.491278886795044,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1075,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5751623511314392,
      "eval_runtime": 1.0114,
      "eval_samples_per_second": 70.2,
      "eval_steps_per_second": 2.966,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5748041868209839,
      "eval_runtime": 0.989,
      "eval_samples_per_second": 71.791,
      "eval_steps_per_second": 3.033,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5776336789131165,
      "eval_runtime": 0.9954,
      "eval_samples_per_second": 71.325,
      "eval_steps_per_second": 3.014,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 2.633176565170288,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.1013,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5844343304634094,
      "eval_runtime": 0.9859,
      "eval_samples_per_second": 72.012,
      "eval_steps_per_second": 3.043,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5929856896400452,
      "eval_runtime": 1.0276,
      "eval_samples_per_second": 69.094,
      "eval_steps_per_second": 2.919,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6019977331161499,
      "eval_runtime": 0.9928,
      "eval_samples_per_second": 71.514,
      "eval_steps_per_second": 3.022,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 3.9098527431488037,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.1092,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6055042147636414,
      "eval_runtime": 1.004,
      "eval_samples_per_second": 70.718,
      "eval_steps_per_second": 2.988,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6074590086936951,
      "eval_runtime": 0.9928,
      "eval_samples_per_second": 71.517,
      "eval_steps_per_second": 3.022,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.607978880405426,
      "eval_runtime": 0.9917,
      "eval_samples_per_second": 71.596,
      "eval_steps_per_second": 3.025,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 2.8199360370635986,
      "learning_rate": 0.0,
      "loss": 0.1096,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6081295013427734,
      "eval_runtime": 1.0121,
      "eval_samples_per_second": 70.154,
      "eval_steps_per_second": 2.964,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 2.8402872494292173e+18,
      "train_loss": 0.24676756938298544,
      "train_runtime": 1640.323,
      "train_samples_per_second": 24.203,
      "train_steps_per_second": 0.183
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.48321232199668884,
      "eval_runtime": 1.0328,
      "eval_samples_per_second": 68.746,
      "eval_steps_per_second": 2.905,
      "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
}