trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9696186166774402,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 1.6767937944408533e-05,
      "loss": 5.8543,
      "step": 500
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.3535875888817068e-05,
      "loss": 0.2421,
      "step": 1000
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.0303813833225597e-05,
      "loss": 0.1757,
      "step": 1500
    },
    {
      "epoch": 0.65,
      "learning_rate": 7.071751777634132e-06,
      "loss": 0.1542,
      "step": 2000
    },
    {
      "epoch": 0.81,
      "learning_rate": 3.839689722042664e-06,
      "loss": 0.1443,
      "step": 2500
    },
    {
      "epoch": 0.97,
      "learning_rate": 6.076276664511959e-07,
      "loss": 0.1375,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3094,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 1.39771592638464e+17,
  "trial_name": null,
  "trial_params": null
}