checkpoint-128/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "eval_steps": 1000,
  "global_step": 128,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.125,
      "grad_norm": 2.4954416751861572,
      "learning_rate": 2.3312246748782624e-05,
      "loss": 6.6234,
      "step": 50
    },
    {
      "epoch": 6.25,
      "grad_norm": 2.288395404815674,
      "learning_rate": 4.466083387394294e-06,
      "loss": 3.4489,
      "step": 100
    }
  ],
  "logging_steps": 50,
  "max_steps": 128,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 1000,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.001
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1138408789377024.0,
  "train_batch_size": 400,
  "trial_name": null,
  "trial_params": null
}