Abstract
Feed-forward layers constitute two-thirds of a transformer model’s parameters, yet their role in the network remains under-explored. We show that feed-forward layers in transformer-based language models operate as key-value memories, where each key correlates with textual patterns in the training examples, and each value induces a distribution over the output vocabulary. Our experiments show that the learned patterns are human-interpretable, and that lower layers tend to capture shallow patterns, while upper layers learn more semantic ones. The values complement the keys’ input patterns by inducing output distributions that concentrate probability mass on tokens likely to appear immediately after each pattern, particularly in the upper layers. Finally, we demonstrate that the output of a feed-forward layer is a composition of its memories, which is subsequently refined throughout the model’s layers via residual connections to produce the final output distribution.- Anthology ID:
- 2021.emnlp-main.446
- Volume:
- Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
- Month:
- November
- Year:
- 2021
- Address:
- Online and Punta Cana, Dominican Republic
- Editors:
- Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
- Venue:
- EMNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5484–5495
- Language:
- URL:
- https://aclanthology.org/2021.emnlp-main.446
- DOI:
- 10.18653/v1/2021.emnlp-main.446
- Cite (ACL):
- Mor Geva, Roei Schuster, Jonathan Berant, and Omer Levy. 2021. Transformer Feed-Forward Layers Are Key-Value Memories. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5484–5495, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
- Cite (Informal):
- Transformer Feed-Forward Layers Are Key-Value Memories (Geva et al., EMNLP 2021)
- PDF:
- https://preview.aclanthology.org/add_acl24_videos/2021.emnlp-main.446.pdf
- Code
- mega002/ff-layers
- Data
- WikiText-103, WikiText-2