English

Exploring Task Performance with Interpretable Models via Sparse Auto-Encoders

Computation and Language 2025-07-10 v1 Machine Learning

Abstract

Large Language Models (LLMs) are traditionally viewed as black-box algorithms, therefore reducing trustworthiness and obscuring potential approaches to increasing performance on downstream tasks. In this work, we apply an effective LLM decomposition method using a dictionary-learning approach with sparse autoencoders. This helps extract monosemantic features from polysemantic LLM neurons. Remarkably, our work identifies model-internal misunderstanding, allowing the automatic reformulation of the prompts with additional annotations to improve the interpretation by LLMs. Moreover, this approach demonstrates a significant performance improvement in downstream tasks, such as mathematical reasoning and metaphor detection.

Keywords

Cite

@article{arxiv.2507.06427,
  title  = {Exploring Task Performance with Interpretable Models via Sparse Auto-Encoders},
  author = {Shun Wang and Tyler Loakman and Youbo Lei and Yi Liu and Bohao Yang and Yuting Zhao and Dong Yang and Chenghua Lin},
  journal= {arXiv preprint arXiv:2507.06427},
  year   = {2025}
}
R2 v1 2026-07-01T03:52:28.047Z