On the Versatility of Sparse Autoencoders for In-Context Learning

Ikhyun Cho, Gaeul Kwon, Julia Hockenmaier


Abstract
Sparse autoencoders (SAEs) are emerging as a key analytical tool in the field of mechanistic interpretability for large language models (LLMs). While SAEs have primarily been used for interpretability, we shift focus and explore an understudied question: “Can SAEs be applied to practical tasks beyond interpretability?” Given that SAEs are trained on billions of tokens for sparse reconstruction, we believe they can serve as effective extractors, offering a wide range of useful knowledge that can benefit practical applications. Building on this motivation, we demonstrate that SAEs can be effectively applied to in-context learning (ICL). In particular, we highlight the utility of the SAE-reconstruction loss by showing that it provides a valuable signal in ICL—exhibiting a strong correlation with LLM performance and offering a powerful unsupervised approach for prompt selection. These findings underscore the versatility of SAEs and reveal their potential for real-world applications beyond interpretability. Our code is available at https://github.com/ihcho2/SAE-GPS.
Anthology ID:
2025.findings-emnlp.1063
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2025
Month:
November
Year:
2025
Address:
Suzhou, China
Editors:
Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
19531–19538
Language:
URL:
https://aclanthology.org/2025.findings-emnlp.1063/
DOI:
Bibkey:
Cite (ACL):
Ikhyun Cho, Gaeul Kwon, and Julia Hockenmaier. 2025. On the Versatility of Sparse Autoencoders for In-Context Learning. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 19531–19538, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
On the Versatility of Sparse Autoencoders for In-Context Learning (Cho et al., Findings 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.findings-emnlp.1063.pdf
Checklist:
 2025.findings-emnlp.1063.checklist.pdf