trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 63,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 183.40280949868244,
      "learning_rate": 4.984280524733107e-07,
      "loss": 5.5793,
      "step": 10
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 163.6354618028223,
      "learning_rate": 4.4545787061700746e-07,
      "loss": 4.3731,
      "step": 20
    },
    {
      "epoch": 1.4390243902439024,
      "grad_norm": 47.7903287681843,
      "learning_rate": 3.325697654887918e-07,
      "loss": 2.1728,
      "step": 30
    },
    {
      "epoch": 1.9268292682926829,
      "grad_norm": 41.00661744940429,
      "learning_rate": 1.9436976651092142e-07,
      "loss": 1.7737,
      "step": 40
    },
    {
      "epoch": 2.3902439024390243,
      "grad_norm": 28.785987188373394,
      "learning_rate": 7.322330470336313e-08,
      "loss": 1.4275,
      "step": 50
    },
    {
      "epoch": 2.8780487804878048,
      "grad_norm": 26.73377169616094,
      "learning_rate": 6.268021954544095e-09,
      "loss": 1.173,
      "step": 60
    }
  ],
  "logging_steps": 10,
  "max_steps": 63,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 16229428625408.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}