Are We Recovering Mechanisms? Objective-Level Recovery Gaps in Mechanistic Interpretability

Chuqin Geng, Li Zhang, Haolin Ye, Mark Zhang, Luke Zhang et al.|October 1, 2026arXiv

Key Takeaway

Standard circuit evaluation metrics can systematically prefer incorrect mechanisms, making better discovery algorithms insufficient without fixing the evaluation objective itself.

Summary

This paper reveals a critical flaw in how mechanistic interpretability evaluates circuit discovery: the standard faithfulness metrics can prefer worse circuits that merely reproduce model behavior without actually capturing the underlying mechanisms.

evaluation safety

Key Terms

mechanistic-interpretability circuit-discovery faithfulness context-distortion attribution-method