trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.5510204081632653,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 1.6521975994110107,
      "eval_runtime": 3389.1646,
      "eval_samples_per_second": 0.784,
      "eval_steps_per_second": 0.098,
      "step": 392
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.1496598639455783e-05,
      "loss": 1.8651,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.5567281246185303,
      "eval_runtime": 3435.2549,
      "eval_samples_per_second": 0.773,
      "eval_steps_per_second": 0.097,
      "step": 784
    },
    {
      "epoch": 2.55,
      "learning_rate": 2.993197278911565e-06,
      "loss": 1.6389,
      "step": 1000
    }
  ],
  "max_steps": 1176,
  "num_train_epochs": 3,
  "total_flos": 1019414568960000.0,
  "trial_name": null,
  "trial_params": null
}