checkpoint-6/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.1764705882352942,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "learning_rate": 7.500000000000001e-05,
      "loss": 2.0533,
      "step": 3
    },
    {
      "epoch": 1.18,
      "learning_rate": 9.333333333333334e-05,
      "loss": 1.5186,
      "step": 6
    }
  ],
  "logging_steps": 3,
  "max_steps": 34,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 787135888097280.0,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}