{
  "best_epoch": 15,
  "best_val_loss": 4.5852992613338595,
  "config": {
    "batch_size": 64,
    "d_model": 225,
    "dropout": 0.1,
    "learning_rate": 0.0001,
    "n_layers": 6,
    "seq_len": 60,
    "state_size": 16
  },
  "final_perplexity": 98.03252009550582,
  "total_epochs": 20,
  "training_duration_hours": 0.021275366113611114
}