Can sparse autoencoders be used to decompose and interpret steering vectors?

🤗 Paper Upvotes: 6 | cs.LG, cs.AI, cs.CL

Authors:
Harry Mayne, Yushi Yang, Adam Mahdi

Title:
Can sparse autoencoders be used to decompose and interpret steering vectors?

Arxiv:
http://arxiv.org/abs/2411.08790v1

Abstract:
Steering vectors are a promising approach to control the behaviour of large language models. However, their underlying mechanisms remain poorly understood. While sparse autoencoders (SAEs) may offer a potential method to interpret steering vectors, recent findings show that SAE-reconstructed vectors often lack the steering properties of the original vectors. This paper investigates why directly applying SAEs to steering vectors yields misleading decompositions, identifying two reasons: (1) steering vectors fall outside the input distribution for which SAEs are designed, and (2) steering vectors can have meaningful negative projections in feature directions, which SAEs are not designed to accommodate. These limitations hinder the direct use of SAEs for interpreting steering vectors.

What is Daily Paper Cast?

We update every weekday to discuss highest-voted papers from Huggingface Daily Paper (https://huggingface.co/papers). Both the podcast scripts and audio are generated by AI. Feedback and suggestions are welcome! Email us: [e-mail]

Creator:
Jingwen Liang, 3D ML, https://www.linkedin.com/in/jingwen-liang/
Gengyu Wang, LLM ML, http://wanggengyu.com

Listen on:
Spotify: https://open.spotify.com/show/21nrhmdaA8qoBiH8q03NXL
Apple Podcast: https://podcasts.apple.com/us/podcast/daily-paper-cast/id1777620236

Cover Image by Kawen Kuang https://kawen.art