trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 1000,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.5,
      "learning_rate": 4.425e-07,
      "loss": 9.0808,
      "step": 25
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.7999999999999996e-07,
      "loss": 8.0507,
      "step": 50
    },
    {
      "epoch": 7.5,
      "learning_rate": 3.175e-07,
      "loss": 7.5948,
      "step": 75
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.55e-07,
      "loss": 7.2855,
      "step": 100
    },
    {
      "epoch": 12.5,
      "learning_rate": 1.9249999999999998e-07,
      "loss": 7.1565,
      "step": 125
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.3e-07,
      "loss": 7.0459,
      "step": 150
    },
    {
      "epoch": 17.5,
      "learning_rate": 6.75e-08,
      "loss": 7.0015,
      "step": 175
    },
    {
      "epoch": 20.0,
      "learning_rate": 5e-09,
      "loss": 6.9769,
      "step": 200
    },
    {
      "epoch": 20.0,
      "step": 200,
      "total_flos": 0.0,
      "train_loss": 7.524070739746094,
      "train_runtime": 153.2004,
      "train_samples_per_second": 1.305,
      "train_steps_per_second": 1.305
    }
  ],
  "logging_steps": 25,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}