last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 4.836557705136758e-05,
      "loss": 0.3773,
      "step": 500
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.08091682195663452,
      "eval_runtime": 6249.7001,
      "eval_samples_per_second": 0.2,
      "eval_steps_per_second": 0.05,
      "step": 500
    }
  ],
  "logging_steps": 500,
  "max_steps": 15000,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2167104602112000.0,
  "trial_name": null,
  "trial_params": null
}