trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 16875,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 7.90167244633686e-05,
      "loss": 1.997,
      "step": 5625
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 103.1554,
      "eval_loss": 1.8580242395401,
      "eval_rouge1": 42.5299,
      "eval_rouge2": 19.1553,
      "eval_rougeL": 28.9273,
      "eval_rougeLsum": 28.9283,
      "eval_runtime": 6558.2204,
      "eval_samples_per_second": 3.05,
      "eval_steps_per_second": 3.05,
      "step": 5625
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.914358456608578e-05,
      "loss": 1.8712,
      "step": 11250
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 104.4534,
      "eval_loss": 1.8425023555755615,
      "eval_rouge1": 42.4591,
      "eval_rouge2": 19.0887,
      "eval_rougeL": 28.9345,
      "eval_rougeLsum": 28.9332,
      "eval_runtime": 6628.4012,
      "eval_samples_per_second": 3.017,
      "eval_steps_per_second": 3.017,
      "step": 11250
    },
    {
      "epoch": 3.0,
      "learning_rate": 5.92686888196304e-05,
      "loss": 1.7596,
      "step": 16875
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 107.2526,
      "eval_loss": 1.8402857780456543,
      "eval_rouge1": 42.5974,
      "eval_rouge2": 19.1909,
      "eval_rougeL": 28.8656,
      "eval_rougeLsum": 28.8652,
      "eval_runtime": 6823.921,
      "eval_samples_per_second": 2.931,
      "eval_steps_per_second": 2.931,
      "step": 16875
    }
  ],
  "max_steps": 50625,
  "num_train_epochs": 9,
  "total_flos": 8.189027929764864e+16,
  "trial_name": null,
  "trial_params": null
}