trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1749,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.57,
      "grad_norm": 2.076580286026001,
      "learning_rate": 1.9499537296421396e-05,
      "loss": 0.7079,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "step": 1749,
      "total_flos": 1.059616188554281e+17,
      "train_loss": 0.6970308891632272,
      "train_runtime": 7601.1179,
      "train_samples_per_second": 0.92,
      "train_steps_per_second": 0.23
    }
  ],
  "logging_steps": 1000,
  "max_steps": 1749,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "total_flos": 1.059616188554281e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}