Steering grids for sparse-autoencoder features: when a top-context label names an activation regime rather than a causal axis
The standard protocol for interpreting sparse-autoencoder (SAE) features labels each feature from its top-activating contexts and validates the label by steering that single feature at a typical magnitude. We argue that this inspects one cell of a larger steering grid, steering condition (single feature, joint feature set, matched random direction) crossed with steering coefficient, and show that other cells carry information that changes the label. On Qwen3-1.7B-Instruct and Gemma-2-2B-it, with the matched-geometry control extended to Llama-3.1-8B-Instruct: (1) features labelled AI self-disclaimer from their top contexts switch to a second surface form under steering, a contemplative voi...
| Authors | Michael A. Riegler, Birk Sebastian Frostelid Torpmann-Hagen |
|---|---|
| Primary category | cs.LG |
| Published | 4 May 2026 |
| Last updated at source | 25 July 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.