{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 40.0,
  "eval_steps": 500,
  "global_step": 1640,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 12.195121951219512,
      "grad_norm": 0.18887688219547272,
      "learning_rate": 2.7829268292682928e-05,
      "loss": 0.1354,
      "step": 500
    },
    {
      "epoch": 24.390243902439025,
      "grad_norm": 0.04567587003111839,
      "learning_rate": 1.5634146341463415e-05,
      "loss": 0.0194,
      "step": 1000
    },
    {
      "epoch": 36.58536585365854,
      "grad_norm": 0.06173429265618324,
      "learning_rate": 3.4390243902439025e-06,
      "loss": 0.0114,
      "step": 1500
    },
    {
      "epoch": 40.0,
      "step": 1640,
      "total_flos": 1.36301316329472e+16,
      "train_loss": 0.0515340986048303,
      "train_runtime": 1191.8255,
      "train_samples_per_second": 43.463,
      "train_steps_per_second": 1.376
    }
  ],
  "logging_steps": 500,
  "max_steps": 1640,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.36301316329472e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}