{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1338,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 1.992537313432836e-05,
      "loss": 2.2911,
      "step": 89
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.8903654485049833e-05,
      "loss": 2.2051,
      "step": 178
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.6686046511627908e-05,
      "loss": 2.1413,
      "step": 267
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.446843853820598e-05,
      "loss": 2.0871,
      "step": 356
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.2250830564784055e-05,
      "loss": 2.0775,
      "step": 445
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.0033222591362127e-05,
      "loss": 2.0572,
      "step": 534
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.78156146179402e-05,
      "loss": 2.0309,
      "step": 623
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.5598006644518274e-05,
      "loss": 2.0246,
      "step": 712
    },
    {
      "epoch": 1.8,
      "learning_rate": 1.3380398671096346e-05,
      "loss": 2.02,
      "step": 801
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.1162790697674418e-05,
      "loss": 2.0198,
      "step": 890
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.945182724252492e-06,
      "loss": 1.9832,
      "step": 979
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.727574750830565e-06,
      "loss": 2.005,
      "step": 1068
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.5099667774086384e-06,
      "loss": 1.9974,
      "step": 1157
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.2923588039867112e-06,
      "loss": 1.9824,
      "step": 1246
    },
    {
      "epoch": 2.99,
      "learning_rate": 7.475083056478404e-08,
      "loss": 1.9968,
      "step": 1335
    }
  ],
  "logging_steps": 89,
  "max_steps": 1338,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.1152005292556288e+17,
  "trial_name": null,
  "trial_params": null
}