trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 1100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 3e-05,
      "loss": 1.4406,
      "step": 275
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 82.35572374645223,
      "eval_f1": 89.7642124543165,
      "eval_runtime": 10.9099,
      "eval_samples_per_second": 974.25,
      "eval_steps_per_second": 3.116,
      "step": 275
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.7644,
      "step": 550
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 83.74645222327341,
      "eval_f1": 90.53757730898367,
      "eval_runtime": 9.4876,
      "eval_samples_per_second": 1120.308,
      "eval_steps_per_second": 3.584,
      "step": 550
    },
    {
      "epoch": 3.0,
      "learning_rate": 3e-05,
      "loss": 0.5654,
      "step": 825
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 82.95175023651845,
      "eval_f1": 90.35323909879975,
      "eval_runtime": 9.4629,
      "eval_samples_per_second": 1123.23,
      "eval_steps_per_second": 3.593,
      "step": 825
    },
    {
      "epoch": 4.0,
      "learning_rate": 3e-05,
      "loss": 0.418,
      "step": 1100
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 83.4720908230842,
      "eval_f1": 90.64073788415168,
      "eval_runtime": 9.4727,
      "eval_samples_per_second": 1122.071,
      "eval_steps_per_second": 3.589,
      "step": 1100
    }
  ],
  "max_steps": 5500,
  "num_train_epochs": 20,
  "total_flos": 348635380318208.0,
  "trial_name": null,
  "trial_params": null
}