{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.32,
  "eval_steps": 500,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 9.972326633975752e-05,
      "loss": 1.6982,
      "step": 50
    },
    {
      "epoch": 0.04,
      "learning_rate": 9.765191054744305e-05,
      "loss": 1.0341,
      "step": 100
    },
    {
      "epoch": 0.06,
      "learning_rate": 9.363468755735123e-05,
      "loss": 1.0017,
      "step": 150
    },
    {
      "epoch": 0.08,
      "learning_rate": 8.783564079088477e-05,
      "loss": 1.0131,
      "step": 200
    },
    {
      "epoch": 0.1,
      "learning_rate": 8.049157449684723e-05,
      "loss": 0.948,
      "step": 250
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.190238384283412e-05,
      "loss": 1.01,
      "step": 300
    },
    {
      "epoch": 0.14,
      "learning_rate": 6.241880868706075e-05,
      "loss": 0.972,
      "step": 350
    },
    {
      "epoch": 0.16,
      "learning_rate": 5.242811110572242e-05,
      "loss": 1.0579,
      "step": 400
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.2338261534486844e-05,
      "loss": 1.0209,
      "step": 450
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.2561279283505883e-05,
      "loss": 0.9866,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 2.349640771651611e-05,
      "loss": 0.9984,
      "step": 550
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.5513811136094787e-05,
      "loss": 1.0059,
      "step": 600
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.939459113251408e-06,
      "loss": 0.9161,
      "step": 650
    },
    {
      "epoch": 0.28,
      "learning_rate": 4.041815510209396e-06,
      "loss": 0.9191,
      "step": 700
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.0208757505326016e-06,
      "loss": 0.9088,
      "step": 750
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0,
      "loss": 0.944,
      "step": 800
    }
  ],
  "logging_steps": 50,
  "max_steps": 800,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 2.48413559980032e+16,
  "trial_name": null,
  "trial_params": null
}