{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 9204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 9.456757931334203e-06,
      "loss": 0.6319,
      "step": 500
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.913515862668406e-06,
      "loss": 0.4731,
      "step": 1000
    },
    {
      "epoch": 0.49,
      "learning_rate": 8.370273794002608e-06,
      "loss": 0.4403,
      "step": 1500
    },
    {
      "epoch": 0.65,
      "learning_rate": 7.827031725336811e-06,
      "loss": 0.4279,
      "step": 2000
    },
    {
      "epoch": 0.81,
      "learning_rate": 7.283789656671013e-06,
      "loss": 0.4164,
      "step": 2500
    },
    {
      "epoch": 0.98,
      "learning_rate": 6.7405475880052155e-06,
      "loss": 0.4081,
      "step": 3000
    },
    {
      "epoch": 1.14,
      "learning_rate": 6.197305519339418e-06,
      "loss": 0.3643,
      "step": 3500
    },
    {
      "epoch": 1.3,
      "learning_rate": 5.65406345067362e-06,
      "loss": 0.3663,
      "step": 4000
    },
    {
      "epoch": 1.47,
      "learning_rate": 5.110821382007823e-06,
      "loss": 0.3533,
      "step": 4500
    },
    {
      "epoch": 1.63,
      "learning_rate": 4.567579313342025e-06,
      "loss": 0.359,
      "step": 5000
    },
    {
      "epoch": 1.79,
      "learning_rate": 4.024337244676228e-06,
      "loss": 0.3517,
      "step": 5500
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.4810951760104307e-06,
      "loss": 0.3553,
      "step": 6000
    },
    {
      "epoch": 2.12,
      "learning_rate": 2.9378531073446333e-06,
      "loss": 0.3249,
      "step": 6500
    },
    {
      "epoch": 2.28,
      "learning_rate": 2.3946110386788354e-06,
      "loss": 0.3133,
      "step": 7000
    },
    {
      "epoch": 2.44,
      "learning_rate": 1.851368970013038e-06,
      "loss": 0.3178,
      "step": 7500
    },
    {
      "epoch": 2.61,
      "learning_rate": 1.3081269013472406e-06,
      "loss": 0.3139,
      "step": 8000
    },
    {
      "epoch": 2.77,
      "learning_rate": 7.648848326814429e-07,
      "loss": 0.3131,
      "step": 8500
    },
    {
      "epoch": 2.93,
      "learning_rate": 2.216427640156454e-07,
      "loss": 0.3173,
      "step": 9000
    },
    {
      "epoch": 3.0,
      "step": 9204,
      "total_flos": 4999000607677440.0,
      "train_runtime": 11284.7294,
      "train_samples_per_second": 0.816
    }
  ],
  "max_steps": 9204,
  "num_train_epochs": 3,
  "total_flos": 4999000607677440.0,
  "trial_name": null,
  "trial_params": null
}