LodeHQ
AI news for people who shipFree, by email

AI · arxiv.org · Monday, 29 June 2026

Hidden Interaction Effects Break Activation Patching’s Causal Estimates

A fresh causal analysis shows activation patching’s natural indirect effect conflates true component influence with interaction effects from other parts of the network. Those hidden interactions can hide or exaggerate a component’s role, calling into question many prior mechanistic interpretability results and suggesting new diagnostic metrics.

Read the source: arxiv.org ↗In the 29 June issue →