trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 366,
  "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": 1.2856,
      "step": 122
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 69.0419717438383,
      "eval_f1": 72.17947350546201,
      "eval_runtime": 4.0061,
      "eval_samples_per_second": 2420.588,
      "eval_steps_per_second": 7.738,
      "step": 122
    },
    {
      "epoch": 2.0,
      "learning_rate": 3e-05,
      "loss": 0.5443,
      "step": 244
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 69.19665876044138,
      "eval_f1": 72.51995998400436,
      "eval_runtime": 3.4927,
      "eval_samples_per_second": 2776.335,
      "eval_steps_per_second": 8.876,
      "step": 244
    },
    {
      "epoch": 3.0,
      "learning_rate": 3e-05,
      "loss": 0.4796,
      "step": 366
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 70.2279055377952,
      "eval_f1": 73.26755492034576,
      "eval_runtime": 3.5334,
      "eval_samples_per_second": 2744.409,
      "eval_steps_per_second": 8.774,
      "step": 366
    }
  ],
  "max_steps": 1220,
  "num_train_epochs": 10,
  "total_flos": 43819446829056.0,
  "trial_name": null,
  "trial_params": null
}