{
  "best_metric": 0.8450704225352113,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-65-fold4/checkpoint-182",
  "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": 0.5211267605633803,
      "eval_loss": 0.7144731879234314,
      "eval_runtime": 1.0188,
      "eval_samples_per_second": 69.692,
      "eval_steps_per_second": 2.945,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.5070422535211268,
      "eval_loss": 0.7082186341285706,
      "eval_runtime": 0.9339,
      "eval_samples_per_second": 76.023,
      "eval_steps_per_second": 3.212,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.6056338028169014,
      "eval_loss": 0.6889063715934753,
      "eval_runtime": 0.9191,
      "eval_samples_per_second": 77.245,
      "eval_steps_per_second": 3.264,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 3.284947633743286,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.6878,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.6703165173530579,
      "eval_runtime": 0.9363,
      "eval_samples_per_second": 75.83,
      "eval_steps_per_second": 3.204,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.6556025743484497,
      "eval_runtime": 0.9344,
      "eval_samples_per_second": 75.984,
      "eval_steps_per_second": 3.211,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.642956554889679,
      "eval_runtime": 0.9474,
      "eval_samples_per_second": 74.941,
      "eval_steps_per_second": 3.167,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 5.985997676849365,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6203,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.6249927282333374,
      "eval_runtime": 0.9539,
      "eval_samples_per_second": 74.435,
      "eval_steps_per_second": 3.145,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6197183098591549,
      "eval_loss": 0.7463871240615845,
      "eval_runtime": 0.9471,
      "eval_samples_per_second": 74.963,
      "eval_steps_per_second": 3.167,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.6056338028169014,
      "eval_loss": 0.6646994352340698,
      "eval_runtime": 0.9245,
      "eval_samples_per_second": 76.797,
      "eval_steps_per_second": 3.245,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 7.882161617279053,
      "learning_rate": 5e-05,
      "loss": 0.5703,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.6096706390380859,
      "eval_runtime": 0.9619,
      "eval_samples_per_second": 73.809,
      "eval_steps_per_second": 3.119,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.6261224746704102,
      "eval_runtime": 0.9396,
      "eval_samples_per_second": 75.56,
      "eval_steps_per_second": 3.193,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.5925809741020203,
      "eval_runtime": 0.9277,
      "eval_samples_per_second": 76.537,
      "eval_steps_per_second": 3.234,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 5.672287940979004,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.5281,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5370251536369324,
      "eval_runtime": 0.9419,
      "eval_samples_per_second": 75.378,
      "eval_steps_per_second": 3.185,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5637533664703369,
      "eval_runtime": 0.9264,
      "eval_samples_per_second": 76.638,
      "eval_steps_per_second": 3.238,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.6056338028169014,
      "eval_loss": 0.7175151109695435,
      "eval_runtime": 0.9684,
      "eval_samples_per_second": 73.32,
      "eval_steps_per_second": 3.098,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 7.855106830596924,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4616,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5774647887323944,
      "eval_loss": 0.8917320966720581,
      "eval_runtime": 0.9446,
      "eval_samples_per_second": 75.161,
      "eval_steps_per_second": 3.176,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.6760779023170471,
      "eval_runtime": 0.9228,
      "eval_samples_per_second": 76.943,
      "eval_steps_per_second": 3.251,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.5606212019920349,
      "eval_runtime": 0.9447,
      "eval_samples_per_second": 75.157,
      "eval_steps_per_second": 3.176,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 5.475687503814697,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.4943,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.6338028169014085,
      "eval_loss": 0.696312427520752,
      "eval_runtime": 0.9291,
      "eval_samples_per_second": 76.418,
      "eval_steps_per_second": 3.229,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.6461552977561951,
      "eval_runtime": 0.9441,
      "eval_samples_per_second": 75.207,
      "eval_steps_per_second": 3.178,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.6246035695075989,
      "eval_runtime": 0.9435,
      "eval_samples_per_second": 75.251,
      "eval_steps_per_second": 3.18,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 4.991897106170654,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.4058,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.6619718309859155,
      "eval_loss": 0.7336124181747437,
      "eval_runtime": 0.9352,
      "eval_samples_per_second": 75.921,
      "eval_steps_per_second": 3.208,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.6270064115524292,
      "eval_runtime": 0.9384,
      "eval_samples_per_second": 75.658,
      "eval_steps_per_second": 3.197,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.6096646189689636,
      "eval_runtime": 0.941,
      "eval_samples_per_second": 75.453,
      "eval_steps_per_second": 3.188,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 4.380879878997803,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.3577,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.6699813008308411,
      "eval_runtime": 0.9378,
      "eval_samples_per_second": 75.708,
      "eval_steps_per_second": 3.199,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.6676053404808044,
      "eval_runtime": 0.9357,
      "eval_samples_per_second": 75.876,
      "eval_steps_per_second": 3.206,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5475191473960876,
      "eval_runtime": 0.9411,
      "eval_samples_per_second": 75.444,
      "eval_steps_per_second": 3.188,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 2.8975579738616943,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2988,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5382919311523438,
      "eval_runtime": 0.9412,
      "eval_samples_per_second": 75.435,
      "eval_steps_per_second": 3.187,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.5533848404884338,
      "eval_runtime": 0.9288,
      "eval_samples_per_second": 76.444,
      "eval_steps_per_second": 3.23,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5842180252075195,
      "eval_runtime": 0.9277,
      "eval_samples_per_second": 76.533,
      "eval_steps_per_second": 3.234,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 3.0538382530212402,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.2595,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5964645743370056,
      "eval_runtime": 0.9103,
      "eval_samples_per_second": 77.994,
      "eval_steps_per_second": 3.296,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.6219843626022339,
      "eval_runtime": 0.9431,
      "eval_samples_per_second": 75.281,
      "eval_steps_per_second": 3.181,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.6026638150215149,
      "eval_runtime": 0.9477,
      "eval_samples_per_second": 74.916,
      "eval_steps_per_second": 3.165,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 3.94685959815979,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.2422,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.6368895769119263,
      "eval_runtime": 0.9216,
      "eval_samples_per_second": 77.044,
      "eval_steps_per_second": 3.255,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6033357381820679,
      "eval_runtime": 0.9401,
      "eval_samples_per_second": 75.526,
      "eval_steps_per_second": 3.191,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.6912439465522766,
      "eval_runtime": 0.9337,
      "eval_samples_per_second": 76.043,
      "eval_steps_per_second": 3.213,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 1.8390007019042969,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1927,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6582470536231995,
      "eval_runtime": 0.9394,
      "eval_samples_per_second": 75.581,
      "eval_steps_per_second": 3.194,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6320064067840576,
      "eval_runtime": 0.9433,
      "eval_samples_per_second": 75.271,
      "eval_steps_per_second": 3.18,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.753158450126648,
      "eval_runtime": 0.9223,
      "eval_samples_per_second": 76.983,
      "eval_steps_per_second": 3.253,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 8.522945404052734,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.2399,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.7909286618232727,
      "eval_runtime": 0.9385,
      "eval_samples_per_second": 75.651,
      "eval_steps_per_second": 3.197,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.6808232665061951,
      "eval_runtime": 0.929,
      "eval_samples_per_second": 76.424,
      "eval_steps_per_second": 3.229,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5815951824188232,
      "eval_runtime": 0.9333,
      "eval_samples_per_second": 76.072,
      "eval_steps_per_second": 3.214,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5473891496658325,
      "eval_runtime": 0.9353,
      "eval_samples_per_second": 75.912,
      "eval_steps_per_second": 3.208,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 3.9758527278900146,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.2218,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.6309952139854431,
      "eval_runtime": 0.9488,
      "eval_samples_per_second": 74.835,
      "eval_steps_per_second": 3.162,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.6452985405921936,
      "eval_runtime": 0.9466,
      "eval_samples_per_second": 75.001,
      "eval_steps_per_second": 3.169,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.616996169090271,
      "eval_runtime": 0.934,
      "eval_samples_per_second": 76.018,
      "eval_steps_per_second": 3.212,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 3.51767635345459,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1817,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.603383481502533,
      "eval_runtime": 0.9385,
      "eval_samples_per_second": 75.653,
      "eval_steps_per_second": 3.197,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6349979043006897,
      "eval_runtime": 0.9472,
      "eval_samples_per_second": 74.957,
      "eval_steps_per_second": 3.167,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6026537418365479,
      "eval_runtime": 0.9329,
      "eval_samples_per_second": 76.103,
      "eval_steps_per_second": 3.216,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 2.8314857482910156,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.1483,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.559914767742157,
      "eval_runtime": 0.9519,
      "eval_samples_per_second": 74.59,
      "eval_steps_per_second": 3.152,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5817492604255676,
      "eval_runtime": 0.9277,
      "eval_samples_per_second": 76.537,
      "eval_steps_per_second": 3.234,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6085928082466125,
      "eval_runtime": 0.9365,
      "eval_samples_per_second": 75.816,
      "eval_steps_per_second": 3.203,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 4.343232154846191,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.1668,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5743552446365356,
      "eval_runtime": 0.9412,
      "eval_samples_per_second": 75.439,
      "eval_steps_per_second": 3.188,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.605885922908783,
      "eval_runtime": 0.9418,
      "eval_samples_per_second": 75.387,
      "eval_steps_per_second": 3.185,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6454694867134094,
      "eval_runtime": 0.9437,
      "eval_samples_per_second": 75.239,
      "eval_steps_per_second": 3.179,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 4.7387166023254395,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1372,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.5367295145988464,
      "eval_runtime": 0.9301,
      "eval_samples_per_second": 76.34,
      "eval_steps_per_second": 3.226,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5615168213844299,
      "eval_runtime": 0.9333,
      "eval_samples_per_second": 76.074,
      "eval_steps_per_second": 3.214,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.6377549767494202,
      "eval_runtime": 0.9271,
      "eval_samples_per_second": 76.581,
      "eval_steps_per_second": 3.236,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 3.1127660274505615,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1485,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5687428116798401,
      "eval_runtime": 0.9342,
      "eval_samples_per_second": 75.999,
      "eval_steps_per_second": 3.211,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.48974475264549255,
      "eval_runtime": 0.9499,
      "eval_samples_per_second": 74.743,
      "eval_steps_per_second": 3.158,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.43841129541397095,
      "eval_runtime": 0.9329,
      "eval_samples_per_second": 76.104,
      "eval_steps_per_second": 3.216,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 2.5722365379333496,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1426,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5086758732795715,
      "eval_runtime": 0.9371,
      "eval_samples_per_second": 75.766,
      "eval_steps_per_second": 3.201,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.47570785880088806,
      "eval_runtime": 0.935,
      "eval_samples_per_second": 75.939,
      "eval_steps_per_second": 3.209,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4372512102127075,
      "eval_runtime": 0.9455,
      "eval_samples_per_second": 75.095,
      "eval_steps_per_second": 3.173,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 3.0248568058013916,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1333,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4512348473072052,
      "eval_runtime": 0.9662,
      "eval_samples_per_second": 73.486,
      "eval_steps_per_second": 3.105,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4619483947753906,
      "eval_runtime": 0.9227,
      "eval_samples_per_second": 76.949,
      "eval_steps_per_second": 3.251,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5519948601722717,
      "eval_runtime": 0.9387,
      "eval_samples_per_second": 75.634,
      "eval_steps_per_second": 3.196,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 3.7715258598327637,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.1306,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5160856246948242,
      "eval_runtime": 0.9414,
      "eval_samples_per_second": 75.417,
      "eval_steps_per_second": 3.187,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5180492401123047,
      "eval_runtime": 0.9311,
      "eval_samples_per_second": 76.252,
      "eval_steps_per_second": 3.222,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5777754783630371,
      "eval_runtime": 0.9362,
      "eval_samples_per_second": 75.843,
      "eval_steps_per_second": 3.205,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 2.9611454010009766,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.1327,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.593269407749176,
      "eval_runtime": 0.9395,
      "eval_samples_per_second": 75.57,
      "eval_steps_per_second": 3.193,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5222498178482056,
      "eval_runtime": 0.9354,
      "eval_samples_per_second": 75.903,
      "eval_steps_per_second": 3.207,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5103858709335327,
      "eval_runtime": 0.941,
      "eval_samples_per_second": 75.455,
      "eval_steps_per_second": 3.188,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 2.8741390705108643,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1171,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5024046301841736,
      "eval_runtime": 0.943,
      "eval_samples_per_second": 75.292,
      "eval_steps_per_second": 3.181,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.50595623254776,
      "eval_runtime": 0.944,
      "eval_samples_per_second": 75.213,
      "eval_steps_per_second": 3.178,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.5267102718353271,
      "eval_runtime": 0.937,
      "eval_samples_per_second": 75.772,
      "eval_steps_per_second": 3.202,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 3.9980766773223877,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1227,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4774617552757263,
      "eval_runtime": 0.9456,
      "eval_samples_per_second": 75.082,
      "eval_steps_per_second": 3.172,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5019908547401428,
      "eval_runtime": 0.9328,
      "eval_samples_per_second": 76.116,
      "eval_steps_per_second": 3.216,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5243141651153564,
      "eval_runtime": 0.9369,
      "eval_samples_per_second": 75.782,
      "eval_steps_per_second": 3.202,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 5.15329647064209,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.1304,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6195448637008667,
      "eval_runtime": 0.9685,
      "eval_samples_per_second": 73.307,
      "eval_steps_per_second": 3.097,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5740300416946411,
      "eval_runtime": 0.9291,
      "eval_samples_per_second": 76.417,
      "eval_steps_per_second": 3.229,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5652071237564087,
      "eval_runtime": 0.9331,
      "eval_samples_per_second": 76.091,
      "eval_steps_per_second": 3.215,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5749574303627014,
      "eval_runtime": 0.9389,
      "eval_samples_per_second": 75.618,
      "eval_steps_per_second": 3.195,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 2.7778007984161377,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1152,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5829400420188904,
      "eval_runtime": 0.9675,
      "eval_samples_per_second": 73.381,
      "eval_steps_per_second": 3.101,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5853610038757324,
      "eval_runtime": 0.9581,
      "eval_samples_per_second": 74.104,
      "eval_steps_per_second": 3.131,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5853627324104309,
      "eval_runtime": 0.9385,
      "eval_samples_per_second": 75.649,
      "eval_steps_per_second": 3.196,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 2.607625722885132,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.1069,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5825924277305603,
      "eval_runtime": 0.9377,
      "eval_samples_per_second": 75.72,
      "eval_steps_per_second": 3.199,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5838791728019714,
      "eval_runtime": 0.9668,
      "eval_samples_per_second": 73.437,
      "eval_steps_per_second": 3.103,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5792059898376465,
      "eval_runtime": 0.9421,
      "eval_samples_per_second": 75.36,
      "eval_steps_per_second": 3.184,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 3.399799108505249,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.122,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5755288600921631,
      "eval_runtime": 0.9561,
      "eval_samples_per_second": 74.263,
      "eval_steps_per_second": 3.138,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5751092433929443,
      "eval_runtime": 0.9263,
      "eval_samples_per_second": 76.65,
      "eval_steps_per_second": 3.239,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5747934579849243,
      "eval_runtime": 0.9473,
      "eval_samples_per_second": 74.952,
      "eval_steps_per_second": 3.167,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 2.51318621635437,
      "learning_rate": 0.0,
      "loss": 0.1268,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5748143196105957,
      "eval_runtime": 0.9381,
      "eval_samples_per_second": 75.682,
      "eval_steps_per_second": 3.198,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 2.8402872494292173e+18,
      "train_loss": 0.25812069336573285,
      "train_runtime": 1594.5895,
      "train_samples_per_second": 24.897,
      "train_steps_per_second": 0.188
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.5367295145988464,
      "eval_runtime": 0.9601,
      "eval_samples_per_second": 73.951,
      "eval_steps_per_second": 3.125,
      "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
}