How does GPT-2 Predict Acronyms? Extracting and Understanding a Circuit via Mechanistic Interpretability

  1. García-Carrasco, J.
  2. Maté, A.
  3. Trujillo, J.
Aktak:
Proceedings of Machine Learning Research

ISSN: 2640-3498

Argitalpen urtea: 2024

Proceedings of the 27th International Conference on Artificial Intelligence and Statistics, AISTATS 2024

Alea: 238

Orrialdeak: 3322-3330

Mota: Biltzar ekarpena