{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 76,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.53,
      "learning_rate": 0.0001736842105263158,
      "loss": 0.9014,
      "step": 10
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.7705,
      "step": 20
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0001236842105263158,
      "loss": 0.6344,
      "step": 30
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.736842105263158e-05,
      "loss": 0.8495,
      "step": 40
    },
    {
      "epoch": 2.63,
      "learning_rate": 7.105263157894737e-05,
      "loss": 0.69,
      "step": 50
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.473684210526316e-05,
      "loss": 0.7047,
      "step": 60
    },
    {
      "epoch": 3.68,
      "learning_rate": 1.8421052631578947e-05,
      "loss": 0.7122,
      "step": 70
    },
    {
      "epoch": 4.0,
      "step": 76,
      "total_flos": 3.243391650443428e+17,
      "train_loss": 0.7333989770788896,
      "train_runtime": 235.128,
      "train_samples_per_second": 5.036,
      "train_steps_per_second": 0.323
    }
  ],
  "max_steps": 76,
  "num_train_epochs": 4,
  "total_flos": 3.243391650443428e+17,
  "trial_name": null,
  "trial_params": null
}