trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.24,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 4.8400000000000004e-05,
      "loss": 2.9182,
      "step": 5
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.64e-05,
      "loss": 2.4304,
      "step": 10
    },
    {
      "epoch": 0.08,
      "eval_loss": 2.1126761436462402,
      "eval_rouge2_fmeasure": 0.2505,
      "eval_rouge2_precision": 0.2538,
      "eval_rouge2_recall": 0.2509,
      "eval_runtime": 13283.7379,
      "eval_samples_per_second": 0.023,
      "eval_steps_per_second": 0.011,
      "step": 10
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.44e-05,
      "loss": 2.2949,
      "step": 15
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.24e-05,
      "loss": 2.145,
      "step": 20
    },
    {
      "epoch": 0.16,
      "eval_loss": 1.9695031642913818,
      "eval_rouge2_fmeasure": 0.2588,
      "eval_rouge2_precision": 0.2769,
      "eval_rouge2_recall": 0.2515,
      "eval_runtime": 13236.2799,
      "eval_samples_per_second": 0.023,
      "eval_steps_per_second": 0.011,
      "step": 20
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.0400000000000006e-05,
      "loss": 2.1084,
      "step": 25
    },
    {
      "epoch": 0.24,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 2.1794,
      "step": 30
    },
    {
      "epoch": 0.24,
      "eval_loss": 1.8257582187652588,
      "eval_rouge2_fmeasure": 0.2789,
      "eval_rouge2_precision": 0.2837,
      "eval_rouge2_recall": 0.2784,
      "eval_runtime": 13267.4385,
      "eval_samples_per_second": 0.023,
      "eval_steps_per_second": 0.011,
      "step": 30
    }
  ],
  "max_steps": 125,
  "num_train_epochs": 1,
  "total_flos": 1296098320711680.0,
  "trial_name": null,
  "trial_params": null
}