trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.06682556300536832,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.98143734360962e-05,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.96287468721924e-05,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.94431203082886e-05,
      "loss": 0.0,
      "step": 1500
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.92574937443848e-05,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.907186718048099e-05,
      "loss": 0.0,
      "step": 2500
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.888624061657719e-05,
      "loss": 0.0,
      "step": 3000
    }
  ],
  "max_steps": 134679,
  "num_train_epochs": 3,
  "total_flos": 6314665328640000.0,
  "trial_name": null,
  "trial_params": null
}