trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8063610851262863,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.47,
      "learning_rate": 4.4153414405986905e-05,
      "loss": 0.6437,
      "step": 500
    },
    {
      "epoch": 0.94,
      "learning_rate": 3.8306828811973814e-05,
      "loss": 0.6233,
      "step": 1000
    },
    {
      "epoch": 1.4,
      "learning_rate": 3.246024321796071e-05,
      "loss": 0.6296,
      "step": 1500
    },
    {
      "epoch": 1.87,
      "learning_rate": 2.6613657623947615e-05,
      "loss": 0.6285,
      "step": 2000
    },
    {
      "epoch": 2.34,
      "learning_rate": 2.076707202993452e-05,
      "loss": 0.6229,
      "step": 2500
    },
    {
      "epoch": 2.81,
      "learning_rate": 1.492048643592142e-05,
      "loss": 0.6255,
      "step": 3000
    }
  ],
  "max_steps": 4276,
  "num_train_epochs": 4,
  "total_flos": 6659303546188800.0,
  "trial_name": null,
  "trial_params": null
}