trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 348,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3.405172413793103e-05,
      "loss": 1.5322,
      "step": 116
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 26.5795,
      "eval_loss": 1.423577070236206,
      "eval_rouge1": 53.9788,
      "eval_rouge2": 29.2977,
      "eval_rougeL": 44.7241,
      "eval_rougeLsum": 49.3707,
      "eval_runtime": 19.5925,
      "eval_samples_per_second": 41.751,
      "eval_steps_per_second": 0.357,
      "step": 116
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.7385057471264368e-05,
      "loss": 1.1154,
      "step": 232
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 29.4071,
      "eval_loss": 1.4339053630828857,
      "eval_rouge1": 54.7261,
      "eval_rouge2": 29.6473,
      "eval_rougeL": 45.0499,
      "eval_rougeLsum": 50.4724,
      "eval_runtime": 20.943,
      "eval_samples_per_second": 39.058,
      "eval_steps_per_second": 0.334,
      "step": 232
    },
    {
      "epoch": 3.0,
      "learning_rate": 7.183908045977011e-07,
      "loss": 0.9569,
      "step": 348
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 30.3435,
      "eval_loss": 1.4327096939086914,
      "eval_rouge1": 54.9942,
      "eval_rouge2": 29.5803,
      "eval_rougeL": 44.8614,
      "eval_rougeLsum": 50.4178,
      "eval_runtime": 20.9993,
      "eval_samples_per_second": 38.954,
      "eval_steps_per_second": 0.333,
      "step": 348
    },
    {
      "epoch": 3.0,
      "step": 348,
      "total_flos": 4.26008990669865e+16,
      "train_loss": 1.2014821151207233,
      "train_runtime": 263.3678,
      "train_samples_per_second": 167.811,
      "train_steps_per_second": 1.321
    }
  ],
  "max_steps": 348,
  "num_train_epochs": 3,
  "total_flos": 4.26008990669865e+16,
  "trial_name": null,
  "trial_params": null
}