{
  "best_metric": 0.4159727096557617,
  "best_model_checkpoint": "vit-msn-small-lateral_flow_ivalidation_train_test_7/checkpoint-3",
  "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": 0.8791208791208791,
      "eval_loss": 0.4159727096557617,
      "eval_runtime": 1.1776,
      "eval_samples_per_second": 231.83,
      "eval_steps_per_second": 4.246,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "eval_accuracy": 0.8388278388278388,
      "eval_loss": 0.4668194353580475,
      "eval_runtime": 1.1816,
      "eval_samples_per_second": 231.05,
      "eval_steps_per_second": 4.232,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "eval_accuracy": 0.8021978021978022,
      "eval_loss": 0.543300986289978,
      "eval_runtime": 1.172,
      "eval_samples_per_second": 232.936,
      "eval_steps_per_second": 4.266,
      "step": 9
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 27.540616989135742,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.3869,
      "step": 10
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8168498168498168,
      "eval_loss": 0.5052347779273987,
      "eval_runtime": 1.166,
      "eval_samples_per_second": 234.13,
      "eval_steps_per_second": 4.288,
      "step": 13
    },
    {
      "epoch": 4.923076923076923,
      "eval_accuracy": 0.8571428571428571,
      "eval_loss": 0.45910125970840454,
      "eval_runtime": 1.1375,
      "eval_samples_per_second": 239.995,
      "eval_steps_per_second": 4.396,
      "step": 16
    },
    {
      "epoch": 5.846153846153846,
      "eval_accuracy": 0.8278388278388278,
      "eval_loss": 0.4820004999637604,
      "eval_runtime": 1.1463,
      "eval_samples_per_second": 238.165,
      "eval_steps_per_second": 4.362,
      "step": 19
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 4.4243292808532715,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.3658,
      "step": 20
    },
    {
      "epoch": 6.769230769230769,
      "eval_accuracy": 0.8095238095238095,
      "eval_loss": 0.4953090250492096,
      "eval_runtime": 1.1575,
      "eval_samples_per_second": 235.858,
      "eval_steps_per_second": 4.32,
      "step": 22
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8608058608058609,
      "eval_loss": 0.4496568441390991,
      "eval_runtime": 1.2191,
      "eval_samples_per_second": 223.934,
      "eval_steps_per_second": 4.101,
      "step": 26
    },
    {
      "epoch": 8.923076923076923,
      "eval_accuracy": 0.8315018315018315,
      "eval_loss": 0.46859756112098694,
      "eval_runtime": 1.1496,
      "eval_samples_per_second": 237.484,
      "eval_steps_per_second": 4.35,
      "step": 29
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 9.303510665893555,
      "learning_rate": 5e-06,
      "loss": 0.3439,
      "step": 30
    },
    {
      "epoch": 9.846153846153847,
      "eval_accuracy": 0.8608058608058609,
      "eval_loss": 0.45060521364212036,
      "eval_runtime": 1.205,
      "eval_samples_per_second": 226.549,
      "eval_steps_per_second": 4.149,
      "step": 32
    },
    {
      "epoch": 10.76923076923077,
      "eval_accuracy": 0.8168498168498168,
      "eval_loss": 0.485858678817749,
      "eval_runtime": 1.1431,
      "eval_samples_per_second": 238.818,
      "eval_steps_per_second": 4.374,
      "step": 35
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8168498168498168,
      "eval_loss": 0.49286317825317383,
      "eval_runtime": 1.1624,
      "eval_samples_per_second": 234.854,
      "eval_steps_per_second": 4.301,
      "step": 39
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 5.916250228881836,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.3416,
      "step": 40
    },
    {
      "epoch": 12.923076923076923,
      "eval_accuracy": 0.8058608058608059,
      "eval_loss": 0.49565577507019043,
      "eval_runtime": 1.2294,
      "eval_samples_per_second": 222.062,
      "eval_steps_per_second": 4.067,
      "step": 42
    },
    {
      "epoch": 13.846153846153847,
      "eval_accuracy": 0.7875457875457875,
      "eval_loss": 0.5228819251060486,
      "eval_runtime": 1.2179,
      "eval_samples_per_second": 224.159,
      "eval_steps_per_second": 4.105,
      "step": 45
    },
    {
      "epoch": 14.76923076923077,
      "eval_accuracy": 0.8534798534798534,
      "eval_loss": 0.44727277755737305,
      "eval_runtime": 1.1724,
      "eval_samples_per_second": 232.862,
      "eval_steps_per_second": 4.265,
      "step": 48
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 8.750184059143066,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.324,
      "step": 50
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8058608058608059,
      "eval_loss": 0.5260118842124939,
      "eval_runtime": 1.1888,
      "eval_samples_per_second": 229.645,
      "eval_steps_per_second": 4.206,
      "step": 52
    },
    {
      "epoch": 16.923076923076923,
      "eval_accuracy": 0.8461538461538461,
      "eval_loss": 0.4581928849220276,
      "eval_runtime": 1.1456,
      "eval_samples_per_second": 238.297,
      "eval_steps_per_second": 4.364,
      "step": 55
    },
    {
      "epoch": 17.846153846153847,
      "eval_accuracy": 0.7838827838827839,
      "eval_loss": 0.5298556089401245,
      "eval_runtime": 1.1808,
      "eval_samples_per_second": 231.203,
      "eval_steps_per_second": 4.234,
      "step": 58
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 11.985779762268066,
      "learning_rate": 1e-05,
      "loss": 0.3273,
      "step": 60
    },
    {
      "epoch": 18.76923076923077,
      "eval_accuracy": 0.8205128205128205,
      "eval_loss": 0.49467265605926514,
      "eval_runtime": 1.2593,
      "eval_samples_per_second": 216.781,
      "eval_steps_per_second": 3.97,
      "step": 61
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7692307692307693,
      "eval_loss": 0.5393444895744324,
      "eval_runtime": 1.2277,
      "eval_samples_per_second": 222.374,
      "eval_steps_per_second": 4.073,
      "step": 65
    },
    {
      "epoch": 20.923076923076923,
      "eval_accuracy": 0.8278388278388278,
      "eval_loss": 0.49159082770347595,
      "eval_runtime": 1.1597,
      "eval_samples_per_second": 235.406,
      "eval_steps_per_second": 4.311,
      "step": 68
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 5.8529887199401855,
      "learning_rate": 9.583333333333335e-06,
      "loss": 0.3397,
      "step": 70
    },
    {
      "epoch": 21.846153846153847,
      "eval_accuracy": 0.7802197802197802,
      "eval_loss": 0.5359878540039062,
      "eval_runtime": 1.1883,
      "eval_samples_per_second": 229.735,
      "eval_steps_per_second": 4.208,
      "step": 71
    },
    {
      "epoch": 22.76923076923077,
      "eval_accuracy": 0.7655677655677655,
      "eval_loss": 0.5660694241523743,
      "eval_runtime": 1.174,
      "eval_samples_per_second": 232.531,
      "eval_steps_per_second": 4.259,
      "step": 74
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.7216117216117216,
      "eval_loss": 0.6353771090507507,
      "eval_runtime": 1.231,
      "eval_samples_per_second": 221.763,
      "eval_steps_per_second": 4.062,
      "step": 78
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 13.490166664123535,
      "learning_rate": 9.166666666666666e-06,
      "loss": 0.3344,
      "step": 80
    },
    {
      "epoch": 24.923076923076923,
      "eval_accuracy": 0.7032967032967034,
      "eval_loss": 0.6782160401344299,
      "eval_runtime": 1.1828,
      "eval_samples_per_second": 230.817,
      "eval_steps_per_second": 4.227,
      "step": 81
    },
    {
      "epoch": 25.846153846153847,
      "eval_accuracy": 0.7582417582417582,
      "eval_loss": 0.5704138875007629,
      "eval_runtime": 1.1635,
      "eval_samples_per_second": 234.632,
      "eval_steps_per_second": 4.297,
      "step": 84
    },
    {
      "epoch": 26.76923076923077,
      "eval_accuracy": 0.6776556776556777,
      "eval_loss": 0.653683066368103,
      "eval_runtime": 1.149,
      "eval_samples_per_second": 237.606,
      "eval_steps_per_second": 4.352,
      "step": 87
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 11.456792831420898,
      "learning_rate": 8.750000000000001e-06,
      "loss": 0.3325,
      "step": 90
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8424908424908425,
      "eval_loss": 0.47976481914520264,
      "eval_runtime": 1.2235,
      "eval_samples_per_second": 223.128,
      "eval_steps_per_second": 4.087,
      "step": 91
    },
    {
      "epoch": 28.923076923076923,
      "eval_accuracy": 0.8058608058608059,
      "eval_loss": 0.5158453583717346,
      "eval_runtime": 1.1703,
      "eval_samples_per_second": 233.277,
      "eval_steps_per_second": 4.272,
      "step": 94
    },
    {
      "epoch": 29.846153846153847,
      "eval_accuracy": 0.7912087912087912,
      "eval_loss": 0.5408114790916443,
      "eval_runtime": 1.2604,
      "eval_samples_per_second": 216.59,
      "eval_steps_per_second": 3.967,
      "step": 97
    },
    {
      "epoch": 30.76923076923077,
      "grad_norm": 13.521626472473145,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.3283,
      "step": 100
    },
    {
      "epoch": 30.76923076923077,
      "eval_accuracy": 0.73992673992674,
      "eval_loss": 0.5963976383209229,
      "eval_runtime": 1.2378,
      "eval_samples_per_second": 220.561,
      "eval_steps_per_second": 4.04,
      "step": 100
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8205128205128205,
      "eval_loss": 0.5069018602371216,
      "eval_runtime": 1.2566,
      "eval_samples_per_second": 217.258,
      "eval_steps_per_second": 3.979,
      "step": 104
    },
    {
      "epoch": 32.92307692307692,
      "eval_accuracy": 0.7875457875457875,
      "eval_loss": 0.5396149754524231,
      "eval_runtime": 1.1548,
      "eval_samples_per_second": 236.412,
      "eval_steps_per_second": 4.33,
      "step": 107
    },
    {
      "epoch": 33.84615384615385,
      "grad_norm": 7.220104694366455,
      "learning_rate": 7.916666666666667e-06,
      "loss": 0.3229,
      "step": 110
    },
    {
      "epoch": 33.84615384615385,
      "eval_accuracy": 0.7985347985347986,
      "eval_loss": 0.52034592628479,
      "eval_runtime": 1.2979,
      "eval_samples_per_second": 210.337,
      "eval_steps_per_second": 3.852,
      "step": 110
    },
    {
      "epoch": 34.76923076923077,
      "eval_accuracy": 0.7875457875457875,
      "eval_loss": 0.5464060306549072,
      "eval_runtime": 1.1496,
      "eval_samples_per_second": 237.467,
      "eval_steps_per_second": 4.349,
      "step": 113
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5890459418296814,
      "eval_runtime": 1.1555,
      "eval_samples_per_second": 236.272,
      "eval_steps_per_second": 4.327,
      "step": 117
    },
    {
      "epoch": 36.92307692307692,
      "grad_norm": 7.553558826446533,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.3207,
      "step": 120
    },
    {
      "epoch": 36.92307692307692,
      "eval_accuracy": 0.8131868131868132,
      "eval_loss": 0.5079773664474487,
      "eval_runtime": 1.1865,
      "eval_samples_per_second": 230.094,
      "eval_steps_per_second": 4.214,
      "step": 120
    },
    {
      "epoch": 37.84615384615385,
      "eval_accuracy": 0.8168498168498168,
      "eval_loss": 0.49443885684013367,
      "eval_runtime": 1.2087,
      "eval_samples_per_second": 225.87,
      "eval_steps_per_second": 4.137,
      "step": 123
    },
    {
      "epoch": 38.76923076923077,
      "eval_accuracy": 0.8095238095238095,
      "eval_loss": 0.49684473872184753,
      "eval_runtime": 1.2281,
      "eval_samples_per_second": 222.303,
      "eval_steps_per_second": 4.071,
      "step": 126
    },
    {
      "epoch": 40.0,
      "grad_norm": 7.015711307525635,
      "learning_rate": 7.083333333333335e-06,
      "loss": 0.3286,
      "step": 130
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8131868131868132,
      "eval_loss": 0.4874400496482849,
      "eval_runtime": 1.1583,
      "eval_samples_per_second": 235.685,
      "eval_steps_per_second": 4.317,
      "step": 130
    },
    {
      "epoch": 40.92307692307692,
      "eval_accuracy": 0.8058608058608059,
      "eval_loss": 0.5012958645820618,
      "eval_runtime": 1.1701,
      "eval_samples_per_second": 233.312,
      "eval_steps_per_second": 4.273,
      "step": 133
    },
    {
      "epoch": 41.84615384615385,
      "eval_accuracy": 0.7655677655677655,
      "eval_loss": 0.5328759551048279,
      "eval_runtime": 1.2067,
      "eval_samples_per_second": 226.236,
      "eval_steps_per_second": 4.144,
      "step": 136
    },
    {
      "epoch": 42.76923076923077,
      "eval_accuracy": 0.6996336996336996,
      "eval_loss": 0.6198887228965759,
      "eval_runtime": 1.1849,
      "eval_samples_per_second": 230.395,
      "eval_steps_per_second": 4.22,
      "step": 139
    },
    {
      "epoch": 43.07692307692308,
      "grad_norm": 12.217916488647461,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.3154,
      "step": 140
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8058608058608059,
      "eval_loss": 0.4853915870189667,
      "eval_runtime": 1.2003,
      "eval_samples_per_second": 227.44,
      "eval_steps_per_second": 4.166,
      "step": 143
    },
    {
      "epoch": 44.92307692307692,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5544903874397278,
      "eval_runtime": 1.2613,
      "eval_samples_per_second": 216.445,
      "eval_steps_per_second": 3.964,
      "step": 146
    },
    {
      "epoch": 45.84615384615385,
      "eval_accuracy": 0.7728937728937729,
      "eval_loss": 0.5267203450202942,
      "eval_runtime": 1.1694,
      "eval_samples_per_second": 233.448,
      "eval_steps_per_second": 4.276,
      "step": 149
    },
    {
      "epoch": 46.15384615384615,
      "grad_norm": 5.860217571258545,
      "learning_rate": 6.25e-06,
      "loss": 0.3119,
      "step": 150
    },
    {
      "epoch": 46.76923076923077,
      "eval_accuracy": 0.7802197802197802,
      "eval_loss": 0.5214048027992249,
      "eval_runtime": 1.1505,
      "eval_samples_per_second": 237.28,
      "eval_steps_per_second": 4.346,
      "step": 152
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.7838827838827839,
      "eval_loss": 0.5265443325042725,
      "eval_runtime": 1.2352,
      "eval_samples_per_second": 221.013,
      "eval_steps_per_second": 4.048,
      "step": 156
    },
    {
      "epoch": 48.92307692307692,
      "eval_accuracy": 0.7985347985347986,
      "eval_loss": 0.5136817693710327,
      "eval_runtime": 1.1863,
      "eval_samples_per_second": 230.125,
      "eval_steps_per_second": 4.215,
      "step": 159
    },
    {
      "epoch": 49.23076923076923,
      "grad_norm": 10.198068618774414,
      "learning_rate": 5.833333333333334e-06,
      "loss": 0.3036,
      "step": 160
    },
    {
      "epoch": 49.84615384615385,
      "eval_accuracy": 0.7838827838827839,
      "eval_loss": 0.5353964567184448,
      "eval_runtime": 1.2084,
      "eval_samples_per_second": 225.92,
      "eval_steps_per_second": 4.138,
      "step": 162
    },
    {
      "epoch": 50.76923076923077,
      "eval_accuracy": 0.7875457875457875,
      "eval_loss": 0.5269169807434082,
      "eval_runtime": 1.1557,
      "eval_samples_per_second": 236.211,
      "eval_steps_per_second": 4.326,
      "step": 165
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.73992673992674,
      "eval_loss": 0.579745888710022,
      "eval_runtime": 1.1971,
      "eval_samples_per_second": 228.046,
      "eval_steps_per_second": 4.177,
      "step": 169
    },
    {
      "epoch": 52.30769230769231,
      "grad_norm": 4.502013206481934,
      "learning_rate": 5.416666666666667e-06,
      "loss": 0.2995,
      "step": 170
    },
    {
      "epoch": 52.92307692307692,
      "eval_accuracy": 0.717948717948718,
      "eval_loss": 0.6257714629173279,
      "eval_runtime": 1.2141,
      "eval_samples_per_second": 224.863,
      "eval_steps_per_second": 4.118,
      "step": 172
    },
    {
      "epoch": 53.84615384615385,
      "eval_accuracy": 0.7692307692307693,
      "eval_loss": 0.5511525869369507,
      "eval_runtime": 1.2158,
      "eval_samples_per_second": 224.549,
      "eval_steps_per_second": 4.113,
      "step": 175
    },
    {
      "epoch": 54.76923076923077,
      "eval_accuracy": 0.7619047619047619,
      "eval_loss": 0.5516501069068909,
      "eval_runtime": 1.1426,
      "eval_samples_per_second": 238.92,
      "eval_steps_per_second": 4.376,
      "step": 178
    },
    {
      "epoch": 55.38461538461539,
      "grad_norm": 7.7625651359558105,
      "learning_rate": 5e-06,
      "loss": 0.306,
      "step": 180
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.7545787545787546,
      "eval_loss": 0.558984637260437,
      "eval_runtime": 1.2053,
      "eval_samples_per_second": 226.503,
      "eval_steps_per_second": 4.148,
      "step": 182
    },
    {
      "epoch": 56.92307692307692,
      "eval_accuracy": 0.7619047619047619,
      "eval_loss": 0.5514388084411621,
      "eval_runtime": 1.1913,
      "eval_samples_per_second": 229.152,
      "eval_steps_per_second": 4.197,
      "step": 185
    },
    {
      "epoch": 57.84615384615385,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5596668124198914,
      "eval_runtime": 1.2098,
      "eval_samples_per_second": 225.659,
      "eval_steps_per_second": 4.133,
      "step": 188
    },
    {
      "epoch": 58.46153846153846,
      "grad_norm": 8.888143539428711,
      "learning_rate": 4.583333333333333e-06,
      "loss": 0.2989,
      "step": 190
    },
    {
      "epoch": 58.76923076923077,
      "eval_accuracy": 0.7326007326007326,
      "eval_loss": 0.5956733822822571,
      "eval_runtime": 1.1927,
      "eval_samples_per_second": 228.893,
      "eval_steps_per_second": 4.192,
      "step": 191
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.7765567765567766,
      "eval_loss": 0.5365986824035645,
      "eval_runtime": 1.2153,
      "eval_samples_per_second": 224.642,
      "eval_steps_per_second": 4.114,
      "step": 195
    },
    {
      "epoch": 60.92307692307692,
      "eval_accuracy": 0.7728937728937729,
      "eval_loss": 0.5464850664138794,
      "eval_runtime": 1.23,
      "eval_samples_per_second": 221.956,
      "eval_steps_per_second": 4.065,
      "step": 198
    },
    {
      "epoch": 61.53846153846154,
      "grad_norm": 10.457231521606445,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.2931,
      "step": 200
    },
    {
      "epoch": 61.84615384615385,
      "eval_accuracy": 0.7252747252747253,
      "eval_loss": 0.6171274781227112,
      "eval_runtime": 1.2391,
      "eval_samples_per_second": 220.329,
      "eval_steps_per_second": 4.035,
      "step": 201
    },
    {
      "epoch": 62.76923076923077,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5768119692802429,
      "eval_runtime": 1.1827,
      "eval_samples_per_second": 230.828,
      "eval_steps_per_second": 4.228,
      "step": 204
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5706220865249634,
      "eval_runtime": 1.1854,
      "eval_samples_per_second": 230.307,
      "eval_steps_per_second": 4.218,
      "step": 208
    },
    {
      "epoch": 64.61538461538461,
      "grad_norm": 6.158288478851318,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.299,
      "step": 210
    },
    {
      "epoch": 64.92307692307692,
      "eval_accuracy": 0.7362637362637363,
      "eval_loss": 0.5962250232696533,
      "eval_runtime": 1.1514,
      "eval_samples_per_second": 237.094,
      "eval_steps_per_second": 4.342,
      "step": 211
    },
    {
      "epoch": 65.84615384615384,
      "eval_accuracy": 0.7216117216117216,
      "eval_loss": 0.6219912767410278,
      "eval_runtime": 1.2105,
      "eval_samples_per_second": 225.52,
      "eval_steps_per_second": 4.13,
      "step": 214
    },
    {
      "epoch": 66.76923076923077,
      "eval_accuracy": 0.7362637362637363,
      "eval_loss": 0.5929316282272339,
      "eval_runtime": 1.1491,
      "eval_samples_per_second": 237.574,
      "eval_steps_per_second": 4.351,
      "step": 217
    },
    {
      "epoch": 67.6923076923077,
      "grad_norm": 5.717153549194336,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.2969,
      "step": 220
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.7252747252747253,
      "eval_loss": 0.6135749816894531,
      "eval_runtime": 1.2314,
      "eval_samples_per_second": 221.695,
      "eval_steps_per_second": 4.06,
      "step": 221
    },
    {
      "epoch": 68.92307692307692,
      "eval_accuracy": 0.7289377289377289,
      "eval_loss": 0.6092182993888855,
      "eval_runtime": 1.165,
      "eval_samples_per_second": 234.334,
      "eval_steps_per_second": 4.292,
      "step": 224
    },
    {
      "epoch": 69.84615384615384,
      "eval_accuracy": 0.7252747252747253,
      "eval_loss": 0.6028769612312317,
      "eval_runtime": 1.1827,
      "eval_samples_per_second": 230.837,
      "eval_steps_per_second": 4.228,
      "step": 227
    },
    {
      "epoch": 70.76923076923077,
      "grad_norm": 4.415896415710449,
      "learning_rate": 2.916666666666667e-06,
      "loss": 0.3015,
      "step": 230
    },
    {
      "epoch": 70.76923076923077,
      "eval_accuracy": 0.7765567765567766,
      "eval_loss": 0.5355977416038513,
      "eval_runtime": 1.2214,
      "eval_samples_per_second": 223.506,
      "eval_steps_per_second": 4.094,
      "step": 230
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.7692307692307693,
      "eval_loss": 0.5375617146492004,
      "eval_runtime": 1.2823,
      "eval_samples_per_second": 212.899,
      "eval_steps_per_second": 3.899,
      "step": 234
    },
    {
      "epoch": 72.92307692307692,
      "eval_accuracy": 0.7435897435897436,
      "eval_loss": 0.5885952115058899,
      "eval_runtime": 1.158,
      "eval_samples_per_second": 235.755,
      "eval_steps_per_second": 4.318,
      "step": 237
    },
    {
      "epoch": 73.84615384615384,
      "grad_norm": 9.932133674621582,
      "learning_rate": 2.5e-06,
      "loss": 0.2919,
      "step": 240
    },
    {
      "epoch": 73.84615384615384,
      "eval_accuracy": 0.7435897435897436,
      "eval_loss": 0.5869150161743164,
      "eval_runtime": 1.2325,
      "eval_samples_per_second": 221.501,
      "eval_steps_per_second": 4.057,
      "step": 240
    },
    {
      "epoch": 74.76923076923077,
      "eval_accuracy": 0.7472527472527473,
      "eval_loss": 0.584571897983551,
      "eval_runtime": 1.1539,
      "eval_samples_per_second": 236.584,
      "eval_steps_per_second": 4.333,
      "step": 243
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.7655677655677655,
      "eval_loss": 0.5507005453109741,
      "eval_runtime": 1.172,
      "eval_samples_per_second": 232.937,
      "eval_steps_per_second": 4.266,
      "step": 247
    },
    {
      "epoch": 76.92307692307692,
      "grad_norm": 8.504337310791016,
      "learning_rate": 2.0833333333333334e-06,
      "loss": 0.288,
      "step": 250
    },
    {
      "epoch": 76.92307692307692,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5800787806510925,
      "eval_runtime": 1.1706,
      "eval_samples_per_second": 233.207,
      "eval_steps_per_second": 4.271,
      "step": 250
    },
    {
      "epoch": 77.84615384615384,
      "eval_accuracy": 0.73992673992674,
      "eval_loss": 0.6076557040214539,
      "eval_runtime": 1.1434,
      "eval_samples_per_second": 238.759,
      "eval_steps_per_second": 4.373,
      "step": 253
    },
    {
      "epoch": 78.76923076923077,
      "eval_accuracy": 0.7435897435897436,
      "eval_loss": 0.5848133563995361,
      "eval_runtime": 1.2128,
      "eval_samples_per_second": 225.096,
      "eval_steps_per_second": 4.123,
      "step": 256
    },
    {
      "epoch": 80.0,
      "grad_norm": 5.167693614959717,
      "learning_rate": 1.6666666666666667e-06,
      "loss": 0.2951,
      "step": 260
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.7692307692307693,
      "eval_loss": 0.5435333847999573,
      "eval_runtime": 1.1913,
      "eval_samples_per_second": 229.17,
      "eval_steps_per_second": 4.197,
      "step": 260
    },
    {
      "epoch": 80.92307692307692,
      "eval_accuracy": 0.7655677655677655,
      "eval_loss": 0.56382155418396,
      "eval_runtime": 1.1629,
      "eval_samples_per_second": 234.755,
      "eval_steps_per_second": 4.3,
      "step": 263
    },
    {
      "epoch": 81.84615384615384,
      "eval_accuracy": 0.73992673992674,
      "eval_loss": 0.5795217156410217,
      "eval_runtime": 1.1592,
      "eval_samples_per_second": 235.498,
      "eval_steps_per_second": 4.313,
      "step": 266
    },
    {
      "epoch": 82.76923076923077,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5773537755012512,
      "eval_runtime": 1.1515,
      "eval_samples_per_second": 237.079,
      "eval_steps_per_second": 4.342,
      "step": 269
    },
    {
      "epoch": 83.07692307692308,
      "grad_norm": 5.818239688873291,
      "learning_rate": 1.25e-06,
      "loss": 0.2875,
      "step": 270
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5702620148658752,
      "eval_runtime": 1.1961,
      "eval_samples_per_second": 228.243,
      "eval_steps_per_second": 4.18,
      "step": 273
    },
    {
      "epoch": 84.92307692307692,
      "eval_accuracy": 0.7509157509157509,
      "eval_loss": 0.5712842345237732,
      "eval_runtime": 1.1634,
      "eval_samples_per_second": 234.664,
      "eval_steps_per_second": 4.298,
      "step": 276
    },
    {
      "epoch": 85.84615384615384,
      "eval_accuracy": 0.7472527472527473,
      "eval_loss": 0.5783692598342896,
      "eval_runtime": 1.204,
      "eval_samples_per_second": 226.74,
      "eval_steps_per_second": 4.153,
      "step": 279
    },
    {
      "epoch": 86.15384615384616,
      "grad_norm": 6.248855113983154,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.2855,
      "step": 280
    },
    {
      "epoch": 86.76923076923077,
      "eval_accuracy": 0.7435897435897436,
      "eval_loss": 0.5903602838516235,
      "eval_runtime": 1.1602,
      "eval_samples_per_second": 235.3,
      "eval_steps_per_second": 4.31,
      "step": 282
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.7326007326007326,
      "eval_loss": 0.5916617512702942,
      "eval_runtime": 1.2119,
      "eval_samples_per_second": 225.264,
      "eval_steps_per_second": 4.126,
      "step": 286
    },
    {
      "epoch": 88.92307692307692,
      "eval_accuracy": 0.7472527472527473,
      "eval_loss": 0.5859794020652771,
      "eval_runtime": 1.1576,
      "eval_samples_per_second": 235.841,
      "eval_steps_per_second": 4.319,
      "step": 289
    },
    {
      "epoch": 89.23076923076923,
      "grad_norm": 5.813708305358887,
      "learning_rate": 4.1666666666666667e-07,
      "loss": 0.2964,
      "step": 290
    },
    {
      "epoch": 89.84615384615384,
      "eval_accuracy": 0.7472527472527473,
      "eval_loss": 0.585824728012085,
      "eval_runtime": 1.2365,
      "eval_samples_per_second": 220.783,
      "eval_steps_per_second": 4.044,
      "step": 292
    },
    {
      "epoch": 90.76923076923077,
      "eval_accuracy": 0.7435897435897436,
      "eval_loss": 0.5823063850402832,
      "eval_runtime": 1.1583,
      "eval_samples_per_second": 235.693,
      "eval_steps_per_second": 4.317,
      "step": 295
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.7435897435897436,
      "eval_loss": 0.5816611647605896,
      "eval_runtime": 1.2737,
      "eval_samples_per_second": 214.329,
      "eval_steps_per_second": 3.925,
      "step": 299
    },
    {
      "epoch": 92.3076923076923,
      "grad_norm": 5.621781826019287,
      "learning_rate": 0.0,
      "loss": 0.291,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "eval_accuracy": 0.7435897435897436,
      "eval_loss": 0.5815560221672058,
      "eval_runtime": 1.2025,
      "eval_samples_per_second": 227.027,
      "eval_steps_per_second": 4.158,
      "step": 300
    },
    {
      "epoch": 92.3076923076923,
      "step": 300,
      "total_flos": 1.4722503891660472e+18,
      "train_loss": 0.31525614658991497,
      "train_runtime": 772.9292,
      "train_samples_per_second": 105.443,
      "train_steps_per_second": 0.388
    }
  ],
  "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": 1.4722503891660472e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}