trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "global_step": 361,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.55,
      "eval_loss": 0.6248018145561218,
      "eval_runtime": 12.7398,
      "eval_samples_per_second": 73.156,
      "eval_steps_per_second": 2.355,
      "step": 200
    },
    {
      "epoch": 1.0,
      "step": 361,
      "total_flos": 2.0319218835442893e+17,
      "train_loss": 1.2307539752315617,
      "train_runtime": 5093.0233,
      "train_samples_per_second": 18.106,
      "train_steps_per_second": 0.071
    }
  ],
  "max_steps": 361,
  "num_train_epochs": 1,
  "total_flos": 2.0319218835442893e+17,
  "trial_name": null,
  "trial_params": null
}