Beyond Activation Patterns: A Weight-Based Out-of-Context Explanation of Sparse Autoencoder Features
This work addresses interpretability for researchers and practitioners using sparse autoencoders in language models, offering a novel out-of-context explanation method.
The paper tackled the problem of interpreting sparse autoencoder features in language models by introducing a weight-based framework that measures functional effects through direct weight interactions, without needing activation data, and found that 1/4 of features directly predict output tokens and features participate in attention mechanisms with distinct profiles.
Sparse autoencoders (SAEs) have emerged as a powerful technique for decomposing language model representations into interpretable features. Current interpretation methods infer feature semantics from activation patterns, but overlook that features are trained to reconstruct activations that serve computational roles in the forward pass. We introduce a novel weight-based interpretation framework that measures functional effects through direct weight interactions, requiring no activation data. Through three experiments on Gemma-2 and Llama-3.1 models, we demonstrate that (1) 1/4 of features directly predict output tokens, (2) features actively participate in attention mechanisms with depth-dependent structure, and (3) semantic and non-semantic feature populations exhibit distinct distribution profiles in attention circuits. Our analysis provides the missing out-of-context half of SAE feature interpretability.