{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 15.961630695443645,
  "eval_steps": 500,
  "global_step": 3328,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_f1": 0.8988,
      "eval_gen_len": 29.90909090909091,
      "eval_loss": 1.8075228929519653,
      "eval_precision": 0.901,
      "eval_recall": 0.897,
      "eval_rouge1": 0.411,
      "eval_rouge2": 0.1689,
      "eval_rougeL": 0.3152,
      "eval_rougeLsum": 0.3155,
      "eval_runtime": 182.3495,
      "eval_samples_per_second": 6.032,
      "eval_steps_per_second": 0.378,
      "step": 208
    },
    {
      "epoch": 2.0,
      "eval_f1": 0.904,
      "eval_gen_len": 29.907272727272726,
      "eval_loss": 1.7311877012252808,
      "eval_precision": 0.9059,
      "eval_recall": 0.9024,
      "eval_rouge1": 0.4379,
      "eval_rouge2": 0.1893,
      "eval_rougeL": 0.3442,
      "eval_rougeLsum": 0.3446,
      "eval_runtime": 160.7204,
      "eval_samples_per_second": 6.844,
      "eval_steps_per_second": 0.429,
      "step": 417
    },
    {
      "epoch": 2.4,
      "learning_rate": 1.699519230769231e-05,
      "loss": 2.0112,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_f1": 0.9055,
      "eval_gen_len": 30.017272727272726,
      "eval_loss": 1.6986640691757202,
      "eval_precision": 0.9075,
      "eval_recall": 0.9039,
      "eval_rouge1": 0.4475,
      "eval_rouge2": 0.1978,
      "eval_rougeL": 0.352,
      "eval_rougeLsum": 0.3525,
      "eval_runtime": 161.6717,
      "eval_samples_per_second": 6.804,
      "eval_steps_per_second": 0.427,
      "step": 625
    },
    {
      "epoch": 4.0,
      "eval_f1": 0.9063,
      "eval_gen_len": 30.061818181818182,
      "eval_loss": 1.676792025566101,
      "eval_precision": 0.9082,
      "eval_recall": 0.9047,
      "eval_rouge1": 0.4514,
      "eval_rouge2": 0.1981,
      "eval_rougeL": 0.357,
      "eval_rougeLsum": 0.3573,
      "eval_runtime": 160.8885,
      "eval_samples_per_second": 6.837,
      "eval_steps_per_second": 0.429,
      "step": 834
    },
    {
      "epoch": 4.8,
      "learning_rate": 1.3990384615384615e-05,
      "loss": 1.7647,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_f1": 0.9068,
      "eval_gen_len": 30.326363636363638,
      "eval_loss": 1.661742091178894,
      "eval_precision": 0.9084,
      "eval_recall": 0.9055,
      "eval_rouge1": 0.4537,
      "eval_rouge2": 0.2003,
      "eval_rougeL": 0.3592,
      "eval_rougeLsum": 0.3595,
      "eval_runtime": 162.5609,
      "eval_samples_per_second": 6.767,
      "eval_steps_per_second": 0.424,
      "step": 1042
    },
    {
      "epoch": 6.0,
      "eval_f1": 0.9072,
      "eval_gen_len": 30.082727272727272,
      "eval_loss": 1.6502095460891724,
      "eval_precision": 0.9089,
      "eval_recall": 0.9057,
      "eval_rouge1": 0.4554,
      "eval_rouge2": 0.2021,
      "eval_rougeL": 0.3607,
      "eval_rougeLsum": 0.361,
      "eval_runtime": 159.1789,
      "eval_samples_per_second": 6.91,
      "eval_steps_per_second": 0.433,
      "step": 1251
    },
    {
      "epoch": 7.0,
      "eval_f1": 0.908,
      "eval_gen_len": 30.021818181818183,
      "eval_loss": 1.6416178941726685,
      "eval_precision": 0.9099,
      "eval_recall": 0.9064,
      "eval_rouge1": 0.4592,
      "eval_rouge2": 0.2052,
      "eval_rougeL": 0.3639,
      "eval_rougeLsum": 0.3641,
      "eval_runtime": 156.6401,
      "eval_samples_per_second": 7.022,
      "eval_steps_per_second": 0.441,
      "step": 1459
    },
    {
      "epoch": 7.19,
      "learning_rate": 1.0985576923076924e-05,
      "loss": 1.6948,
      "step": 1500
    },
    {
      "epoch": 8.0,
      "eval_f1": 0.9081,
      "eval_gen_len": 30.78272727272727,
      "eval_loss": 1.6360372304916382,
      "eval_precision": 0.909,
      "eval_recall": 0.9074,
      "eval_rouge1": 0.4612,
      "eval_rouge2": 0.2054,
      "eval_rougeL": 0.3649,
      "eval_rougeLsum": 0.365,
      "eval_runtime": 160.7238,
      "eval_samples_per_second": 6.844,
      "eval_steps_per_second": 0.429,
      "step": 1668
    },
    {
      "epoch": 9.0,
      "eval_f1": 0.9083,
      "eval_gen_len": 30.62909090909091,
      "eval_loss": 1.6301532983779907,
      "eval_precision": 0.9095,
      "eval_recall": 0.9074,
      "eval_rouge1": 0.4621,
      "eval_rouge2": 0.2062,
      "eval_rougeL": 0.3645,
      "eval_rougeLsum": 0.3647,
      "eval_runtime": 159.8662,
      "eval_samples_per_second": 6.881,
      "eval_steps_per_second": 0.432,
      "step": 1876
    },
    {
      "epoch": 9.59,
      "learning_rate": 7.980769230769232e-06,
      "loss": 1.6501,
      "step": 2000
    },
    {
      "epoch": 10.0,
      "eval_f1": 0.9083,
      "eval_gen_len": 30.48181818181818,
      "eval_loss": 1.6264721155166626,
      "eval_precision": 0.9095,
      "eval_recall": 0.9073,
      "eval_rouge1": 0.4606,
      "eval_rouge2": 0.2051,
      "eval_rougeL": 0.3651,
      "eval_rougeLsum": 0.3655,
      "eval_runtime": 157.2711,
      "eval_samples_per_second": 6.994,
      "eval_steps_per_second": 0.439,
      "step": 2085
    },
    {
      "epoch": 11.0,
      "eval_f1": 0.9087,
      "eval_gen_len": 30.806363636363635,
      "eval_loss": 1.6229554414749146,
      "eval_precision": 0.9097,
      "eval_recall": 0.908,
      "eval_rouge1": 0.4625,
      "eval_rouge2": 0.2073,
      "eval_rougeL": 0.3658,
      "eval_rougeLsum": 0.366,
      "eval_runtime": 157.1263,
      "eval_samples_per_second": 7.001,
      "eval_steps_per_second": 0.439,
      "step": 2293
    },
    {
      "epoch": 11.99,
      "learning_rate": 4.975961538461539e-06,
      "loss": 1.6222,
      "step": 2500
    },
    {
      "epoch": 12.0,
      "eval_f1": 0.909,
      "eval_gen_len": 30.55272727272727,
      "eval_loss": 1.6204967498779297,
      "eval_precision": 0.9103,
      "eval_recall": 0.9081,
      "eval_rouge1": 0.4644,
      "eval_rouge2": 0.2082,
      "eval_rougeL": 0.3674,
      "eval_rougeLsum": 0.3679,
      "eval_runtime": 155.7494,
      "eval_samples_per_second": 7.063,
      "eval_steps_per_second": 0.443,
      "step": 2502
    },
    {
      "epoch": 13.0,
      "eval_f1": 0.909,
      "eval_gen_len": 30.805454545454545,
      "eval_loss": 1.618751883506775,
      "eval_precision": 0.9101,
      "eval_recall": 0.9083,
      "eval_rouge1": 0.4648,
      "eval_rouge2": 0.2087,
      "eval_rougeL": 0.3681,
      "eval_rougeLsum": 0.3683,
      "eval_runtime": 158.59,
      "eval_samples_per_second": 6.936,
      "eval_steps_per_second": 0.435,
      "step": 2710
    },
    {
      "epoch": 14.0,
      "eval_f1": 0.9093,
      "eval_gen_len": 30.670909090909092,
      "eval_loss": 1.6172130107879639,
      "eval_precision": 0.9104,
      "eval_recall": 0.9084,
      "eval_rouge1": 0.4654,
      "eval_rouge2": 0.2097,
      "eval_rougeL": 0.3685,
      "eval_rougeLsum": 0.3689,
      "eval_runtime": 159.3516,
      "eval_samples_per_second": 6.903,
      "eval_steps_per_second": 0.433,
      "step": 2919
    },
    {
      "epoch": 14.39,
      "learning_rate": 1.971153846153846e-06,
      "loss": 1.6048,
      "step": 3000
    },
    {
      "epoch": 15.0,
      "eval_f1": 0.9093,
      "eval_gen_len": 30.630909090909093,
      "eval_loss": 1.6168792247772217,
      "eval_precision": 0.9104,
      "eval_recall": 0.9084,
      "eval_rouge1": 0.465,
      "eval_rouge2": 0.21,
      "eval_rougeL": 0.3693,
      "eval_rougeLsum": 0.3697,
      "eval_runtime": 158.4169,
      "eval_samples_per_second": 6.944,
      "eval_steps_per_second": 0.436,
      "step": 3127
    },
    {
      "epoch": 15.96,
      "eval_f1": 0.9091,
      "eval_gen_len": 30.619090909090907,
      "eval_loss": 1.6166560649871826,
      "eval_precision": 0.9102,
      "eval_recall": 0.9083,
      "eval_rouge1": 0.4649,
      "eval_rouge2": 0.2096,
      "eval_rougeL": 0.3686,
      "eval_rougeLsum": 0.3688,
      "eval_runtime": 157.7317,
      "eval_samples_per_second": 6.974,
      "eval_steps_per_second": 0.437,
      "step": 3328
    },
    {
      "epoch": 15.96,
      "step": 3328,
      "total_flos": 4.6122414295331635e+17,
      "train_loss": 1.711962864949153,
      "train_runtime": 11907.5247,
      "train_samples_per_second": 26.874,
      "train_steps_per_second": 0.279
    }
  ],
  "logging_steps": 500,
  "max_steps": 3328,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 16,
  "save_steps": 500,
  "total_flos": 4.6122414295331635e+17,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}