{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 4620,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.08,
      "learning_rate": 4.458874458874459e-05,
      "loss": 0.4026,
      "step": 500
    },
    {
      "epoch": 2.16,
      "learning_rate": 3.917748917748918e-05,
      "loss": 0.1339,
      "step": 1000
    },
    {
      "epoch": 3.25,
      "learning_rate": 3.376623376623377e-05,
      "loss": 0.066,
      "step": 1500
    },
    {
      "epoch": 4.33,
      "learning_rate": 2.8354978354978357e-05,
      "loss": 0.032,
      "step": 2000
    },
    {
      "epoch": 5.41,
      "learning_rate": 2.2943722943722946e-05,
      "loss": 0.0157,
      "step": 2500
    },
    {
      "epoch": 6.49,
      "learning_rate": 1.7532467532467535e-05,
      "loss": 0.0061,
      "step": 3000
    },
    {
      "epoch": 7.58,
      "learning_rate": 1.2121212121212122e-05,
      "loss": 0.0051,
      "step": 3500
    },
    {
      "epoch": 8.66,
      "learning_rate": 6.709956709956711e-06,
      "loss": 0.002,
      "step": 4000
    },
    {
      "epoch": 9.74,
      "learning_rate": 1.2987012987012988e-06,
      "loss": 0.0015,
      "step": 4500
    },
    {
      "epoch": 10.0,
      "step": 4620,
      "total_flos": 2411508200294400.0,
      "train_loss": 0.07197833191200252,
      "train_runtime": 2030.3618,
      "train_samples_per_second": 18.179,
      "train_steps_per_second": 2.275
    }
  ],
  "max_steps": 4620,
  "num_train_epochs": 10,
  "total_flos": 2411508200294400.0,
  "trial_name": null,
  "trial_params": null
}