checkpoint-500/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5333333333333333,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 1.715923547744751,
      "learning_rate": 0.00027878464818763325,
      "loss": 2.0994,
      "step": 200
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 1.554486870765686,
      "learning_rate": 0.00025746268656716415,
      "loss": 1.7202,
      "step": 400
    }
  ],
  "logging_steps": 200,
  "max_steps": 2814,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 541367205888000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}