Knowledge editing experiments may mislead interpretability conclusions

yoavgo · x · 2026-08-26

Yoav Golberger argues that some seemingly causal or actionable results can mislead interpretability conclusions. Citing 'knowledge editing' as an example, he notes that editing weights at one location and observing a behavior change does not prove knowledge is stored there. The intervention likely targets the last node in a path, not necessarily the storage location. While this constitutes good actionable research, it is poor interpretability research.

Related event: Yoav Goldberg Sparks Debate: Mechanistic Understanding Should Trump Steering in Interpretability Research(8 posts)→

Original post →

More from Research

Research channel →