paper-with-me

Papers

Improving Steering Vectors by Targeting Sparse Autoencoder Features

2024-11-04 · Sviatoslav Chalnev, Matthew Siu, Arthur Conmy

To control the behavior of language models, steering methods attempt to ensure that outputs of the model satisfy specific pre-defined properties. Adding steering vectors to the model is a promising method of model control that is easier than finetuning, and may be more robust than prompting. However, it can be difficult to anticipate the effects of steering vectors produced by methods such as CAA [Panickssery et al., 2024] or the direct use of SAE latents [Templeton et al., 2024]. In our work, we address this issue by using SAEs to measure the effects of steering vectors, giving us a method that can be used to understand the causal effect of any steering vector intervention. We use this method for measuring causal effects to develop an improved steering method, SAE-Targeted Steering (SAE-TS), which finds steering vectors to target specific SAE features while minimizing unintended side effects. We show that overall, SAE-TS balances steering effects with coherence better than CAA and SAE feature steering, when evaluated on a range of tasks.

📄 PDF Abstract BibTeX arXiv:2411.02193

Code (1)

slavachalnev/sae-ts 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Denoising Concept Vectors with Sparse Autoencoders for Improved Language Model Steering

2025-05-21 · Haiyan Zhao, Xuansheng Wu, Fan Yang, Bo Shen 외

Linear Concept Vectors have proven effective for steering large language models (LLMs). While existing approaches like linear probing and difference-in-means derive these vectors from LLM hidden representations, diverse …

counterfactualDenoisingLanguage ModelingLanguage Modelling

Enhancing LLM Steering through Sparse Autoencoder-Based Vector Refinement

2025-09-28 · Anyi Wang, Xuansheng Wu, Dong Shu, Yunpu Ma 외 arxiv

Steering has emerged as a promising approach in controlling large language models (LLMs) without modifying model parameters. However, most existing steering methods rely on large-scale datasets to learn clear behavioral …

Semantic Similarity

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

2024-11-13 · Harry Mayne, Yushi Yang, Adam Mahdi

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…

Steering Large Language Models with Feature Guided Activation Additions

2025-01-17 · Samuel Soo, Wesley Teng, Chandrasekaran Balaganesh

Effective and reliable control over large language model (LLM) behavior is a significant challenge. While activation steering methods, which add steering vectors to a model's hidden states, are a promising approach, exis…

DecoderLanguage ModelingLanguage ModellingLarge Language Model

Graph-Regularized Sparse Autoencoders for LLM Safety Steering

2025-12-07 · Jehyeok Yeon, Federico Cinus, Yifan Wu, Luca Luceri arxiv

Sparse autoencoders (SAEs) are increasingly used to extract activation directions for inference-time steering, but their standard sparsity objective treats latent features as independent. This prior can be poorly matched…