{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 111,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 1.5531,
      "step": 7
    },
    {
      "epoch": 0.38,
      "learning_rate": 2.9393939393939394e-05,
      "loss": 1.4741,
      "step": 14
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 1.364,
      "step": 21
    },
    {
      "epoch": 0.76,
      "learning_rate": 2.5151515151515152e-05,
      "loss": 1.2773,
      "step": 28
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.303030303030303e-05,
      "loss": 1.0824,
      "step": 35
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.090909090909091e-05,
      "loss": 1.0063,
      "step": 42
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.878787878787879e-05,
      "loss": 0.9659,
      "step": 49
    },
    {
      "epoch": 1.51,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.9725,
      "step": 56
    },
    {
      "epoch": 1.7,
      "learning_rate": 1.4545454545454546e-05,
      "loss": 0.9026,
      "step": 63
    },
    {
      "epoch": 1.89,
      "learning_rate": 1.2424242424242425e-05,
      "loss": 0.8159,
      "step": 70
    },
    {
      "epoch": 2.08,
      "learning_rate": 1.0303030303030302e-05,
      "loss": 0.8412,
      "step": 77
    },
    {
      "epoch": 2.27,
      "learning_rate": 8.181818181818181e-06,
      "loss": 0.9272,
      "step": 84
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.060606060606061e-06,
      "loss": 0.7917,
      "step": 91
    },
    {
      "epoch": 2.65,
      "learning_rate": 3.93939393939394e-06,
      "loss": 0.8091,
      "step": 98
    },
    {
      "epoch": 2.84,
      "learning_rate": 2.121212121212121e-06,
      "loss": 0.8021,
      "step": 105
    }
  ],
  "logging_steps": 7,
  "max_steps": 111,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 9576819971850240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}