SharedSAE: One Feature Dictionary Across Language Models

2026-09-03

AI Summary

研究提出SharedSAE方法，通过共享特征字典和模型特定的编码器-解码器对来统一不同语言模型的稀疏自动编码器，以保留激活幅度并提高跨模型的一致性和效率。

Abstract

Sparse autoencoders (SAEs) are widely used to interpret language model activations, but SAE training and latent labelling are typically repeated for every model. Here, we show that a single shared SAE can replace a collection of dedicated per-model SAEs. Our method, SharedSAE, combines a shared dictionary with model-specific encoder-decoder pairs. Unlike the closest prior method, which discards activation magnitudes and requires all models at inference, SharedSAE instead normalizes only selection scores, preserving magnitudes, and uses model dropout for single-model inference. We train SharedSAE on four 1B-scale base language models spanning distinct families and tokenizers. Despite sharing its latents across models, SharedSAE retains 96.6% of dedicated SAEs' mean explained variance; its latent activations exhibit cross-model correlations 1.8 times as high as separate SAEs aligned post-hoc, and its latent descriptions transfer across models. After the dictionary is frozen, new models can be efficiently adapted to it, achieving near-dedicated-SAE reconstruction quality while reusing the shared latent descriptions.

Problem

• Sparse Autoencoders
• Language Models
• Shared Dictionary

Innovation

• SharedSAE
• shared dictionary
• model-specific encoders-decoders
• activation magnitude preservation
• model dropout