trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.319148936170213,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 132.7733,
      "eval_loss": 1.3755453824996948,
      "eval_rouge1": 53.8464,
      "eval_rouge2": 25.6231,
      "eval_rougeL": 36.8116,
      "eval_rougeLsum": 49.9957,
      "eval_runtime": 155.2181,
      "eval_samples_per_second": 0.966,
      "eval_steps_per_second": 0.245,
      "step": 94
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 137.1267,
      "eval_loss": 1.2081164121627808,
      "eval_rouge1": 55.5247,
      "eval_rouge2": 27.5408,
      "eval_rougeL": 37.7199,
      "eval_rougeLsum": 51.7518,
      "eval_runtime": 169.6036,
      "eval_samples_per_second": 0.884,
      "eval_steps_per_second": 0.224,
      "step": 188
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 137.62,
      "eval_loss": 1.1149141788482666,
      "eval_rouge1": 55.6412,
      "eval_rouge2": 28.4076,
      "eval_rougeL": 39.0516,
      "eval_rougeLsum": 52.4525,
      "eval_runtime": 161.1061,
      "eval_samples_per_second": 0.931,
      "eval_steps_per_second": 0.236,
      "step": 282
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 135.1733,
      "eval_loss": 1.0563586950302124,
      "eval_rouge1": 56.5376,
      "eval_rouge2": 29.8226,
      "eval_rougeL": 39.9324,
      "eval_rougeLsum": 53.2425,
      "eval_runtime": 157.7547,
      "eval_samples_per_second": 0.951,
      "eval_steps_per_second": 0.241,
      "step": 376
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 135.1733,
      "eval_loss": 1.0107210874557495,
      "eval_rouge1": 57.7194,
      "eval_rouge2": 31.6091,
      "eval_rougeL": 41.9389,
      "eval_rougeLsum": 54.465,
      "eval_runtime": 160.0847,
      "eval_samples_per_second": 0.937,
      "eval_steps_per_second": 0.237,
      "step": 470
    },
    {
      "epoch": 5.32,
      "learning_rate": 9.361702127659576e-06,
      "loss": 1.1609,
      "step": 500
    }
  ],
  "max_steps": 940,
  "num_train_epochs": 10,
  "total_flos": 4323373681213440.0,
  "trial_name": null,
  "trial_params": null
}