trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 258,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 116.65116279069767,
      "eval_loss": 1.3532294034957886,
      "eval_rouge1": 54.8564,
      "eval_rouge2": 29.5263,
      "eval_rougeL": 36.6465,
      "eval_rougeLsum": 50.2558,
      "eval_runtime": 36.2335,
      "eval_samples_per_second": 2.373,
      "eval_steps_per_second": 0.607,
      "step": 86
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 117.94186046511628,
      "eval_loss": 1.3117806911468506,
      "eval_rouge1": 56.6239,
      "eval_rouge2": 31.6121,
      "eval_rougeL": 39.2945,
      "eval_rougeLsum": 51.7651,
      "eval_runtime": 36.4324,
      "eval_samples_per_second": 2.361,
      "eval_steps_per_second": 0.604,
      "step": 172
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 119.45348837209302,
      "eval_loss": 1.3380043506622314,
      "eval_rouge1": 56.6424,
      "eval_rouge2": 31.6294,
      "eval_rougeL": 38.8938,
      "eval_rougeLsum": 51.9078,
      "eval_runtime": 37.5694,
      "eval_samples_per_second": 2.289,
      "eval_steps_per_second": 0.586,
      "step": 258
    },
    {
      "epoch": 3.0,
      "step": 258,
      "total_flos": 2083493304631296.0,
      "train_loss": 1.2955651098443557,
      "train_runtime": 223.6512,
      "train_samples_per_second": 4.588,
      "train_steps_per_second": 1.154
    }
  ],
  "logging_steps": 500,
  "max_steps": 258,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2083493304631296.0,
  "trial_name": null,
  "trial_params": null
}