{
  "best_metric": 0.9064748201438849,
  "best_model_checkpoint": "train_authorship/train_outputs/05-26-2024_17:09:07/checkpoint-300",
  "epoch": 10.0,
  "eval_steps": 15,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "grad_norm": 5.808341979980469,
      "learning_rate": 2.5e-05,
      "loss": 1.0972,
      "step": 15
    },
    {
      "epoch": 0.5,
      "eval_acc_product": 0.21928799282400072,
      "eval_bush_acc": 0.8201438848920863,
      "eval_loss": 0.9504616856575012,
      "eval_obama_acc": 0.45323741007194246,
      "eval_overall_acc": 0.6211031175059952,
      "eval_runtime": 2.8206,
      "eval_samples_per_second": 147.84,
      "eval_steps_per_second": 2.482,
      "eval_trump_acc": 0.5899280575539568,
      "step": 15
    },
    {
      "epoch": 1.0,
      "grad_norm": 48.525230407714844,
      "learning_rate": 5e-05,
      "loss": 0.8423,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_acc_product": 0.39412887680642716,
      "eval_bush_acc": 0.5755395683453237,
      "eval_loss": 0.648549497127533,
      "eval_obama_acc": 0.7266187050359713,
      "eval_overall_acc": 0.7482014388489209,
      "eval_runtime": 2.7763,
      "eval_samples_per_second": 150.202,
      "eval_steps_per_second": 2.521,
      "eval_trump_acc": 0.9424460431654677,
      "step": 30
    },
    {
      "epoch": 1.5,
      "grad_norm": 12.380006790161133,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.5581,
      "step": 45
    },
    {
      "epoch": 1.5,
      "eval_acc_product": 0.43508777678442107,
      "eval_bush_acc": 0.9640287769784173,
      "eval_loss": 0.6018202304840088,
      "eval_obama_acc": 0.5755395683453237,
      "eval_overall_acc": 0.7745803357314148,
      "eval_runtime": 2.7813,
      "eval_samples_per_second": 149.93,
      "eval_steps_per_second": 2.517,
      "eval_trump_acc": 0.7841726618705036,
      "step": 45
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.6562857627868652,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.4719,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_acc_product": 0.5783158370565594,
      "eval_bush_acc": 0.7482014388489209,
      "eval_loss": 0.4394747018814087,
      "eval_obama_acc": 0.8201438848920863,
      "eval_overall_acc": 0.8369304556354916,
      "eval_runtime": 2.7839,
      "eval_samples_per_second": 149.787,
      "eval_steps_per_second": 2.514,
      "eval_trump_acc": 0.9424460431654677,
      "step": 60
    },
    {
      "epoch": 2.5,
      "grad_norm": 14.938467025756836,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.265,
      "step": 75
    },
    {
      "epoch": 2.5,
      "eval_acc_product": 0.5558941905013333,
      "eval_bush_acc": 0.7913669064748201,
      "eval_loss": 0.4741947650909424,
      "eval_obama_acc": 0.8345323741007195,
      "eval_overall_acc": 0.8225419664268585,
      "eval_runtime": 2.7828,
      "eval_samples_per_second": 149.851,
      "eval_steps_per_second": 2.515,
      "eval_trump_acc": 0.841726618705036,
      "step": 75
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4905509948730469,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.1729,
      "step": 90
    },
    {
      "epoch": 3.0,
      "eval_acc_product": 0.25067517023077357,
      "eval_bush_acc": 0.302158273381295,
      "eval_loss": 1.0451148748397827,
      "eval_obama_acc": 0.841726618705036,
      "eval_overall_acc": 0.709832134292566,
      "eval_runtime": 2.7825,
      "eval_samples_per_second": 149.863,
      "eval_steps_per_second": 2.516,
      "eval_trump_acc": 0.9856115107913669,
      "step": 90
    },
    {
      "epoch": 3.5,
      "grad_norm": 32.64374542236328,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.2945,
      "step": 105
    },
    {
      "epoch": 3.5,
      "eval_acc_product": 0.5622189893652079,
      "eval_bush_acc": 0.7338129496402878,
      "eval_loss": 0.5390242338180542,
      "eval_obama_acc": 0.8129496402877698,
      "eval_overall_acc": 0.829736211031175,
      "eval_runtime": 2.7882,
      "eval_samples_per_second": 149.56,
      "eval_steps_per_second": 2.511,
      "eval_trump_acc": 0.9424460431654677,
      "step": 105
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.29498815536499023,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1925,
      "step": 120
    },
    {
      "epoch": 4.0,
      "eval_acc_product": 0.6885265557028006,
      "eval_bush_acc": 0.935251798561151,
      "eval_loss": 0.42252907156944275,
      "eval_obama_acc": 0.8057553956834532,
      "eval_overall_acc": 0.8848920863309353,
      "eval_runtime": 2.7847,
      "eval_samples_per_second": 149.745,
      "eval_steps_per_second": 2.514,
      "eval_trump_acc": 0.9136690647482014,
      "step": 120
    },
    {
      "epoch": 4.5,
      "grad_norm": 5.147099018096924,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.0396,
      "step": 135
    },
    {
      "epoch": 4.5,
      "eval_acc_product": 0.628604429742268,
      "eval_bush_acc": 0.920863309352518,
      "eval_loss": 0.6601312160491943,
      "eval_obama_acc": 0.8705035971223022,
      "eval_overall_acc": 0.8585131894484412,
      "eval_runtime": 2.7927,
      "eval_samples_per_second": 149.316,
      "eval_steps_per_second": 2.507,
      "eval_trump_acc": 0.7841726618705036,
      "step": 135
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.057328637689352036,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.0438,
      "step": 150
    },
    {
      "epoch": 5.0,
      "eval_acc_product": 0.6582616521554249,
      "eval_bush_acc": 0.8633093525179856,
      "eval_loss": 0.6852461695671082,
      "eval_obama_acc": 0.9136690647482014,
      "eval_overall_acc": 0.8705035971223022,
      "eval_runtime": 2.7951,
      "eval_samples_per_second": 149.19,
      "eval_steps_per_second": 2.504,
      "eval_trump_acc": 0.8345323741007195,
      "step": 150
    },
    {
      "epoch": 5.5,
      "grad_norm": 1.9609147310256958,
      "learning_rate": 2.5e-05,
      "loss": 0.0264,
      "step": 165
    },
    {
      "epoch": 5.5,
      "eval_acc_product": 0.6410678506519354,
      "eval_bush_acc": 0.9064748201438849,
      "eval_loss": 0.7376511096954346,
      "eval_obama_acc": 0.8057553956834532,
      "eval_overall_acc": 0.8633093525179856,
      "eval_runtime": 2.7923,
      "eval_samples_per_second": 149.337,
      "eval_steps_per_second": 2.507,
      "eval_trump_acc": 0.8776978417266187,
      "step": 165
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.07944349199533463,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.0444,
      "step": 180
    },
    {
      "epoch": 6.0,
      "eval_acc_product": 0.6344280406118664,
      "eval_bush_acc": 0.9280575539568345,
      "eval_loss": 0.775315523147583,
      "eval_obama_acc": 0.7482014388489209,
      "eval_overall_acc": 0.8633093525179856,
      "eval_runtime": 2.7918,
      "eval_samples_per_second": 149.364,
      "eval_steps_per_second": 2.507,
      "eval_trump_acc": 0.9136690647482014,
      "step": 180
    },
    {
      "epoch": 6.5,
      "grad_norm": 17.292938232421875,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.0344,
      "step": 195
    },
    {
      "epoch": 6.5,
      "eval_acc_product": 0.6043388879807597,
      "eval_bush_acc": 0.9784172661870504,
      "eval_loss": 0.862628698348999,
      "eval_obama_acc": 0.7338129496402878,
      "eval_overall_acc": 0.8513189448441247,
      "eval_runtime": 2.7928,
      "eval_samples_per_second": 149.314,
      "eval_steps_per_second": 2.506,
      "eval_trump_acc": 0.841726618705036,
      "step": 195
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.018421683460474014,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.0185,
      "step": 210
    },
    {
      "epoch": 7.0,
      "eval_acc_product": 0.698349244624796,
      "eval_bush_acc": 0.8992805755395683,
      "eval_loss": 0.6285017132759094,
      "eval_obama_acc": 0.8920863309352518,
      "eval_overall_acc": 0.8872901678657075,
      "eval_runtime": 2.7927,
      "eval_samples_per_second": 149.321,
      "eval_steps_per_second": 2.507,
      "eval_trump_acc": 0.8705035971223022,
      "step": 210
    },
    {
      "epoch": 7.5,
      "grad_norm": 0.12802745401859283,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.0114,
      "step": 225
    },
    {
      "epoch": 7.5,
      "eval_acc_product": 0.6621914724314953,
      "eval_bush_acc": 0.8129496402877698,
      "eval_loss": 0.7886965870857239,
      "eval_obama_acc": 0.9280575539568345,
      "eval_overall_acc": 0.8729016786570744,
      "eval_runtime": 2.7919,
      "eval_samples_per_second": 149.361,
      "eval_steps_per_second": 2.507,
      "eval_trump_acc": 0.8776978417266187,
      "step": 225
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.008947780355811119,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0025,
      "step": 240
    },
    {
      "epoch": 8.0,
      "eval_acc_product": 0.7039360385817944,
      "eval_bush_acc": 0.9064748201438849,
      "eval_loss": 0.6208213567733765,
      "eval_obama_acc": 0.8920863309352518,
      "eval_overall_acc": 0.8896882494004796,
      "eval_runtime": 2.7916,
      "eval_samples_per_second": 149.375,
      "eval_steps_per_second": 2.508,
      "eval_trump_acc": 0.8705035971223022,
      "step": 240
    },
    {
      "epoch": 8.5,
      "grad_norm": 0.06868788599967957,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0069,
      "step": 255
    },
    {
      "epoch": 8.5,
      "eval_acc_product": 0.7387000911149348,
      "eval_bush_acc": 0.9136690647482014,
      "eval_loss": 0.6432462334632874,
      "eval_obama_acc": 0.9136690647482014,
      "eval_overall_acc": 0.9040767386091128,
      "eval_runtime": 2.796,
      "eval_samples_per_second": 149.139,
      "eval_steps_per_second": 2.504,
      "eval_trump_acc": 0.8848920863309353,
      "step": 255
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.00421088607981801,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0019,
      "step": 270
    },
    {
      "epoch": 9.0,
      "eval_acc_product": 0.7271128183111604,
      "eval_bush_acc": 0.8848920863309353,
      "eval_loss": 0.6422414779663086,
      "eval_obama_acc": 0.9064748201438849,
      "eval_overall_acc": 0.8992805755395683,
      "eval_runtime": 2.794,
      "eval_samples_per_second": 149.248,
      "eval_steps_per_second": 2.505,
      "eval_trump_acc": 0.9064748201438849,
      "step": 270
    },
    {
      "epoch": 9.5,
      "grad_norm": 0.2265014946460724,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.0009,
      "step": 285
    },
    {
      "epoch": 9.5,
      "eval_acc_product": 0.7271128183111604,
      "eval_bush_acc": 0.8848920863309353,
      "eval_loss": 0.6394339799880981,
      "eval_obama_acc": 0.9064748201438849,
      "eval_overall_acc": 0.8992805755395683,
      "eval_runtime": 2.7933,
      "eval_samples_per_second": 149.288,
      "eval_steps_per_second": 2.506,
      "eval_trump_acc": 0.9064748201438849,
      "step": 285
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.013048585504293442,
      "learning_rate": 0.0,
      "loss": 0.0021,
      "step": 300
    },
    {
      "epoch": 10.0,
      "eval_acc_product": 0.7448472772943594,
      "eval_bush_acc": 0.9064748201438849,
      "eval_loss": 0.6265907287597656,
      "eval_obama_acc": 0.9064748201438849,
      "eval_overall_acc": 0.9064748201438849,
      "eval_runtime": 2.7927,
      "eval_samples_per_second": 149.318,
      "eval_steps_per_second": 2.507,
      "eval_trump_acc": 0.9064748201438849,
      "step": 300
    }
  ],
  "logging_steps": 15,
  "max_steps": 300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 15,
  "total_flos": 5170662082636590.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}