{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 12.5,
      "learning_rate": 4.375e-05,
      "loss": 0.8815,
      "step": 500
    },
    {
      "epoch": 25.0,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.0559,
      "step": 1000
    },
    {
      "epoch": 37.5,
      "learning_rate": 3.125e-05,
      "loss": 0.0193,
      "step": 1500
    },
    {
      "epoch": 50.0,
      "learning_rate": 2.5e-05,
      "loss": 0.009,
      "step": 2000
    },
    {
      "epoch": 62.5,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.0058,
      "step": 2500
    },
    {
      "epoch": 75.0,
      "learning_rate": 1.25e-05,
      "loss": 0.0042,
      "step": 3000
    },
    {
      "epoch": 87.5,
      "learning_rate": 6.25e-06,
      "loss": 0.0027,
      "step": 3500
    },
    {
      "epoch": 100.0,
      "learning_rate": 0.0,
      "loss": 0.002,
      "step": 4000
    },
    {
      "epoch": 100.0,
      "step": 4000,
      "total_flos": 4.8994912512e+16,
      "train_loss": 0.12255857245624066,
      "train_runtime": 2635.091,
      "train_samples_per_second": 189.747,
      "train_steps_per_second": 1.518
    }
  ],
  "logging_steps": 500,
  "max_steps": 4000,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 4.8994912512e+16,
  "trial_name": null,
  "trial_params": null
}