{
  "best_metric": 0.8732394366197183,
  "best_model_checkpoint": "beit-base-patch16-224-65-fold4/checkpoint-156",
  "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": 0.5352112676056338,
      "eval_loss": 0.7414544224739075,
      "eval_runtime": 1.1337,
      "eval_samples_per_second": 62.625,
      "eval_steps_per_second": 2.646,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.4507042253521127,
      "eval_loss": 0.7177426815032959,
      "eval_runtime": 1.1244,
      "eval_samples_per_second": 63.144,
      "eval_steps_per_second": 2.668,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.6056338028169014,
      "eval_loss": 0.6709167957305908,
      "eval_runtime": 1.0896,
      "eval_samples_per_second": 65.163,
      "eval_steps_per_second": 2.753,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 10.040596008300781,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.748,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6338028169014085,
      "eval_loss": 0.6332956552505493,
      "eval_runtime": 1.0739,
      "eval_samples_per_second": 66.112,
      "eval_steps_per_second": 2.793,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.7323943661971831,
      "eval_loss": 0.616184651851654,
      "eval_runtime": 1.1084,
      "eval_samples_per_second": 64.056,
      "eval_steps_per_second": 2.707,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.6338028169014085,
      "eval_loss": 0.6302564144134521,
      "eval_runtime": 1.0932,
      "eval_samples_per_second": 64.948,
      "eval_steps_per_second": 2.744,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 10.696013450622559,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6397,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.676056338028169,
      "eval_loss": 0.5950036644935608,
      "eval_runtime": 1.1254,
      "eval_samples_per_second": 63.09,
      "eval_steps_per_second": 2.666,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6056338028169014,
      "eval_loss": 0.6324568390846252,
      "eval_runtime": 1.0982,
      "eval_samples_per_second": 64.652,
      "eval_steps_per_second": 2.732,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.57989501953125,
      "eval_runtime": 1.1111,
      "eval_samples_per_second": 63.903,
      "eval_steps_per_second": 2.7,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 3.6701440811157227,
      "learning_rate": 5e-05,
      "loss": 0.5957,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.5793185234069824,
      "eval_runtime": 1.1242,
      "eval_samples_per_second": 63.154,
      "eval_steps_per_second": 2.668,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.7183098591549296,
      "eval_loss": 0.5868809819221497,
      "eval_runtime": 1.0792,
      "eval_samples_per_second": 65.788,
      "eval_steps_per_second": 2.78,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5774647887323944,
      "eval_loss": 0.6194613575935364,
      "eval_runtime": 1.0651,
      "eval_samples_per_second": 66.66,
      "eval_steps_per_second": 2.817,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 12.420059204101562,
      "learning_rate": 4.814814814814815e-05,
      "loss": 0.5676,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.647887323943662,
      "eval_loss": 0.594038188457489,
      "eval_runtime": 1.0942,
      "eval_samples_per_second": 64.89,
      "eval_steps_per_second": 2.742,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.6197183098591549,
      "eval_loss": 0.66124427318573,
      "eval_runtime": 1.1074,
      "eval_samples_per_second": 64.115,
      "eval_steps_per_second": 2.709,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5597983002662659,
      "eval_runtime": 1.1062,
      "eval_samples_per_second": 64.186,
      "eval_steps_per_second": 2.712,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 7.216217994689941,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.5952,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5471624135971069,
      "eval_runtime": 1.123,
      "eval_samples_per_second": 63.226,
      "eval_steps_per_second": 2.672,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.4822922945022583,
      "eval_runtime": 1.1012,
      "eval_samples_per_second": 64.476,
      "eval_steps_per_second": 2.724,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.6901408450704225,
      "eval_loss": 0.6493080854415894,
      "eval_runtime": 1.1156,
      "eval_samples_per_second": 63.642,
      "eval_steps_per_second": 2.689,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 12.423552513122559,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.4908,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.5539372563362122,
      "eval_runtime": 1.1082,
      "eval_samples_per_second": 64.067,
      "eval_steps_per_second": 2.707,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5406185984611511,
      "eval_runtime": 1.1084,
      "eval_samples_per_second": 64.057,
      "eval_steps_per_second": 2.707,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.5443481206893921,
      "eval_runtime": 1.0706,
      "eval_samples_per_second": 66.317,
      "eval_steps_per_second": 2.802,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 4.6347761154174805,
      "learning_rate": 4.259259259259259e-05,
      "loss": 0.4474,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.704225352112676,
      "eval_loss": 0.6547633409500122,
      "eval_runtime": 1.1171,
      "eval_samples_per_second": 63.56,
      "eval_steps_per_second": 2.686,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.4923730790615082,
      "eval_runtime": 1.0861,
      "eval_samples_per_second": 65.372,
      "eval_steps_per_second": 2.762,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.4671289622783661,
      "eval_runtime": 1.1176,
      "eval_samples_per_second": 63.529,
      "eval_steps_per_second": 2.684,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 4.023049354553223,
      "learning_rate": 4.074074074074074e-05,
      "loss": 0.4106,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.4116646349430084,
      "eval_runtime": 1.0944,
      "eval_samples_per_second": 64.878,
      "eval_steps_per_second": 2.741,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.46300172805786133,
      "eval_runtime": 1.1005,
      "eval_samples_per_second": 64.516,
      "eval_steps_per_second": 2.726,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.49149656295776367,
      "eval_runtime": 1.0821,
      "eval_samples_per_second": 65.616,
      "eval_steps_per_second": 2.772,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 8.567398071289062,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.3163,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.633648157119751,
      "eval_runtime": 1.0778,
      "eval_samples_per_second": 65.873,
      "eval_steps_per_second": 2.783,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5920482277870178,
      "eval_runtime": 1.11,
      "eval_samples_per_second": 63.962,
      "eval_steps_per_second": 2.703,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5652734637260437,
      "eval_runtime": 1.0992,
      "eval_samples_per_second": 64.592,
      "eval_steps_per_second": 2.729,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 8.468196868896484,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.3234,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6410530805587769,
      "eval_runtime": 1.1131,
      "eval_samples_per_second": 63.785,
      "eval_steps_per_second": 2.695,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6728390455245972,
      "eval_runtime": 1.1019,
      "eval_samples_per_second": 64.432,
      "eval_steps_per_second": 2.722,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.5502663850784302,
      "eval_runtime": 1.1118,
      "eval_samples_per_second": 63.863,
      "eval_steps_per_second": 2.698,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 4.566059589385986,
      "learning_rate": 3.518518518518519e-05,
      "loss": 0.2969,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.49144864082336426,
      "eval_runtime": 1.0811,
      "eval_samples_per_second": 65.677,
      "eval_steps_per_second": 2.775,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5952464938163757,
      "eval_runtime": 1.0822,
      "eval_samples_per_second": 65.61,
      "eval_steps_per_second": 2.772,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7161255478858948,
      "eval_runtime": 1.0926,
      "eval_samples_per_second": 64.983,
      "eval_steps_per_second": 2.746,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 6.610856533050537,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.2325,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6517188549041748,
      "eval_runtime": 1.0913,
      "eval_samples_per_second": 65.062,
      "eval_steps_per_second": 2.749,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.5832222104072571,
      "eval_runtime": 1.0839,
      "eval_samples_per_second": 65.501,
      "eval_steps_per_second": 2.768,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.6309386491775513,
      "eval_runtime": 1.1093,
      "eval_samples_per_second": 64.005,
      "eval_steps_per_second": 2.704,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 7.450739860534668,
      "learning_rate": 3.148148148148148e-05,
      "loss": 0.2447,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.7464788732394366,
      "eval_loss": 0.8010722994804382,
      "eval_runtime": 1.095,
      "eval_samples_per_second": 64.838,
      "eval_steps_per_second": 2.74,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6084908246994019,
      "eval_runtime": 1.0953,
      "eval_samples_per_second": 64.822,
      "eval_steps_per_second": 2.739,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.6470026969909668,
      "eval_runtime": 1.1169,
      "eval_samples_per_second": 63.567,
      "eval_steps_per_second": 2.686,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.7743595838546753,
      "eval_runtime": 1.1188,
      "eval_samples_per_second": 63.46,
      "eval_steps_per_second": 2.681,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 7.353163719177246,
      "learning_rate": 2.962962962962963e-05,
      "loss": 0.2217,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5730376243591309,
      "eval_runtime": 1.0965,
      "eval_samples_per_second": 64.749,
      "eval_steps_per_second": 2.736,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.5577320456504822,
      "eval_runtime": 1.1035,
      "eval_samples_per_second": 64.34,
      "eval_steps_per_second": 2.719,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.8450704225352113,
      "eval_loss": 0.5226030945777893,
      "eval_runtime": 1.0942,
      "eval_samples_per_second": 64.889,
      "eval_steps_per_second": 2.742,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 4.83720588684082,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.2231,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5114508867263794,
      "eval_runtime": 1.0938,
      "eval_samples_per_second": 64.914,
      "eval_steps_per_second": 2.743,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.5414801836013794,
      "eval_runtime": 1.1045,
      "eval_samples_per_second": 64.284,
      "eval_steps_per_second": 2.716,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.5971347093582153,
      "eval_runtime": 1.1056,
      "eval_samples_per_second": 64.219,
      "eval_steps_per_second": 2.713,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 4.444674015045166,
      "learning_rate": 2.5925925925925925e-05,
      "loss": 0.2014,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.8717265129089355,
      "eval_runtime": 1.0857,
      "eval_samples_per_second": 65.394,
      "eval_steps_per_second": 2.763,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.706287682056427,
      "eval_runtime": 1.088,
      "eval_samples_per_second": 65.257,
      "eval_steps_per_second": 2.757,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6917403936386108,
      "eval_runtime": 1.0985,
      "eval_samples_per_second": 64.636,
      "eval_steps_per_second": 2.731,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 4.06027889251709,
      "learning_rate": 2.4074074074074074e-05,
      "loss": 0.1827,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.6879878044128418,
      "eval_runtime": 1.0966,
      "eval_samples_per_second": 64.745,
      "eval_steps_per_second": 2.736,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7026765942573547,
      "eval_runtime": 1.1021,
      "eval_samples_per_second": 64.422,
      "eval_steps_per_second": 2.722,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.676358699798584,
      "eval_runtime": 1.1066,
      "eval_samples_per_second": 64.158,
      "eval_steps_per_second": 2.711,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 5.703511714935303,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1558,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.739784300327301,
      "eval_runtime": 1.1035,
      "eval_samples_per_second": 64.342,
      "eval_steps_per_second": 2.719,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7787145376205444,
      "eval_runtime": 1.0836,
      "eval_samples_per_second": 65.524,
      "eval_steps_per_second": 2.769,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.767788290977478,
      "eval_runtime": 1.0956,
      "eval_samples_per_second": 64.804,
      "eval_steps_per_second": 2.738,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 4.812498569488525,
      "learning_rate": 2.037037037037037e-05,
      "loss": 0.1637,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.7605633802816901,
      "eval_loss": 0.7898280024528503,
      "eval_runtime": 1.1051,
      "eval_samples_per_second": 64.246,
      "eval_steps_per_second": 2.715,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7105039954185486,
      "eval_runtime": 1.1088,
      "eval_samples_per_second": 64.031,
      "eval_steps_per_second": 2.706,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.8591549295774648,
      "eval_loss": 0.7261584401130676,
      "eval_runtime": 1.1004,
      "eval_samples_per_second": 64.52,
      "eval_steps_per_second": 2.726,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 4.3177595138549805,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1591,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.746425986289978,
      "eval_runtime": 1.148,
      "eval_samples_per_second": 61.845,
      "eval_steps_per_second": 2.613,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.723250150680542,
      "eval_runtime": 1.0977,
      "eval_samples_per_second": 64.68,
      "eval_steps_per_second": 2.733,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7263011932373047,
      "eval_runtime": 1.0978,
      "eval_samples_per_second": 64.672,
      "eval_steps_per_second": 2.733,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 6.147719383239746,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1521,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7376906871795654,
      "eval_runtime": 1.1001,
      "eval_samples_per_second": 64.54,
      "eval_steps_per_second": 2.727,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7266830205917358,
      "eval_runtime": 1.104,
      "eval_samples_per_second": 64.313,
      "eval_steps_per_second": 2.717,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7177650332450867,
      "eval_runtime": 1.1001,
      "eval_samples_per_second": 64.542,
      "eval_steps_per_second": 2.727,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 6.513650894165039,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.157,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.8584705591201782,
      "eval_runtime": 1.1173,
      "eval_samples_per_second": 63.545,
      "eval_steps_per_second": 2.685,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.8628968596458435,
      "eval_runtime": 1.0874,
      "eval_samples_per_second": 65.294,
      "eval_steps_per_second": 2.759,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.7329018115997314,
      "eval_runtime": 1.0893,
      "eval_samples_per_second": 65.18,
      "eval_steps_per_second": 2.754,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 5.656296253204346,
      "learning_rate": 1.2962962962962962e-05,
      "loss": 0.1593,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.6996561884880066,
      "eval_runtime": 1.1005,
      "eval_samples_per_second": 64.518,
      "eval_steps_per_second": 2.726,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.807439386844635,
      "eval_runtime": 1.1052,
      "eval_samples_per_second": 64.24,
      "eval_steps_per_second": 2.714,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 1.0351884365081787,
      "eval_runtime": 1.076,
      "eval_samples_per_second": 65.983,
      "eval_steps_per_second": 2.788,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 8.042988777160645,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.134,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 1.047208309173584,
      "eval_runtime": 1.0895,
      "eval_samples_per_second": 65.17,
      "eval_steps_per_second": 2.754,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.8169014084507042,
      "eval_loss": 0.7477359771728516,
      "eval_runtime": 1.1144,
      "eval_samples_per_second": 63.709,
      "eval_steps_per_second": 2.692,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7356883883476257,
      "eval_runtime": 1.1434,
      "eval_samples_per_second": 62.094,
      "eval_steps_per_second": 2.624,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 4.592105865478516,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.1386,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.8496927618980408,
      "eval_runtime": 1.111,
      "eval_samples_per_second": 63.904,
      "eval_steps_per_second": 2.7,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.7746478873239436,
      "eval_loss": 0.9463661909103394,
      "eval_runtime": 1.1042,
      "eval_samples_per_second": 64.297,
      "eval_steps_per_second": 2.717,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.7887323943661971,
      "eval_loss": 0.8534867763519287,
      "eval_runtime": 1.1034,
      "eval_samples_per_second": 64.347,
      "eval_steps_per_second": 2.719,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 6.3139801025390625,
      "learning_rate": 7.4074074074074075e-06,
      "loss": 0.1246,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.7998347282409668,
      "eval_runtime": 1.096,
      "eval_samples_per_second": 64.781,
      "eval_steps_per_second": 2.737,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.8309859154929577,
      "eval_loss": 0.8214295506477356,
      "eval_runtime": 1.0962,
      "eval_samples_per_second": 64.772,
      "eval_steps_per_second": 2.737,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8373579382896423,
      "eval_runtime": 1.0984,
      "eval_samples_per_second": 64.639,
      "eval_steps_per_second": 2.731,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8597092032432556,
      "eval_runtime": 1.0776,
      "eval_samples_per_second": 65.89,
      "eval_steps_per_second": 2.784,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 4.384837627410889,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.1271,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8436772227287292,
      "eval_runtime": 1.0988,
      "eval_samples_per_second": 64.618,
      "eval_steps_per_second": 2.73,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8370133638381958,
      "eval_runtime": 1.1195,
      "eval_samples_per_second": 63.422,
      "eval_steps_per_second": 2.68,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8297713398933411,
      "eval_runtime": 1.1173,
      "eval_samples_per_second": 63.548,
      "eval_steps_per_second": 2.685,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 5.161994457244873,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.1274,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8339681625366211,
      "eval_runtime": 1.0995,
      "eval_samples_per_second": 64.576,
      "eval_steps_per_second": 2.729,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8462348580360413,
      "eval_runtime": 1.1041,
      "eval_samples_per_second": 64.307,
      "eval_steps_per_second": 2.717,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8593801259994507,
      "eval_runtime": 1.0721,
      "eval_samples_per_second": 66.222,
      "eval_steps_per_second": 2.798,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 5.839167594909668,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.1251,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8504469394683838,
      "eval_runtime": 1.0936,
      "eval_samples_per_second": 64.924,
      "eval_steps_per_second": 2.743,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8480309247970581,
      "eval_runtime": 1.1078,
      "eval_samples_per_second": 64.091,
      "eval_steps_per_second": 2.708,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8470619320869446,
      "eval_runtime": 1.1163,
      "eval_samples_per_second": 63.602,
      "eval_steps_per_second": 2.687,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 4.478762149810791,
      "learning_rate": 0.0,
      "loss": 0.1207,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8028169014084507,
      "eval_loss": 0.8469257354736328,
      "eval_runtime": 1.0756,
      "eval_samples_per_second": 66.01,
      "eval_steps_per_second": 2.789,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 2.839022453308834e+18,
      "train_loss": 0.28607154568036397,
      "train_runtime": 1730.6786,
      "train_samples_per_second": 22.939,
      "train_steps_per_second": 0.173
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.8732394366197183,
      "eval_loss": 0.5414801836013794,
      "eval_runtime": 1.0783,
      "eval_samples_per_second": 65.843,
      "eval_steps_per_second": 2.782,
      "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
}