checkpoint-2048/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.27809084119763733,
  "eval_steps": 500,
  "global_step": 2048,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13904542059881866,
      "grad_norm": 0.6612324714660645,
      "learning_rate": 4.9951171875e-05,
      "loss": 0.7928,
      "step": 1024
    },
    {
      "epoch": 0.27809084119763733,
      "grad_norm": 0.350844144821167,
      "learning_rate": 4.193344898281029e-05,
      "loss": 0.2139,
      "step": 2048
    }
  ],
  "logging_steps": 1024,
  "max_steps": 7365,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2048,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.913440784224354e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}