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": 108.24418604651163,
      "eval_loss": 1.6037122011184692,
      "eval_rouge1": 51.0086,
      "eval_rouge2": 27.7443,
      "eval_rougeL": 35.7708,
      "eval_rougeLsum": 47.0373,
      "eval_runtime": 22.9482,
      "eval_samples_per_second": 3.748,
      "eval_steps_per_second": 0.959,
      "step": 86
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 109.11627906976744,
      "eval_loss": 1.5580047369003296,
      "eval_rouge1": 53.4388,
      "eval_rouge2": 29.9039,
      "eval_rougeL": 37.6105,
      "eval_rougeLsum": 49.6511,
      "eval_runtime": 22.5751,
      "eval_samples_per_second": 3.81,
      "eval_steps_per_second": 0.975,
      "step": 172
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 106.6046511627907,
      "eval_loss": 1.5455105304718018,
      "eval_rouge1": 53.0097,
      "eval_rouge2": 29.8987,
      "eval_rougeL": 37.7495,
      "eval_rougeLsum": 49.1744,
      "eval_runtime": 22.4771,
      "eval_samples_per_second": 3.826,
      "eval_steps_per_second": 0.979,
      "step": 258
    },
    {
      "epoch": 3.0,
      "step": 258,
      "total_flos": 583436809912320.0,
      "train_loss": 1.6933694292408552,
      "train_runtime": 107.1989,
      "train_samples_per_second": 9.571,
      "train_steps_per_second": 2.407
    }
  ],
  "logging_steps": 500,
  "max_steps": 258,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 583436809912320.0,
  "trial_name": null,
  "trial_params": null
}