How to use and interpret activation patching

Activation patching is a popular <PRE_TAG>mechanistic interpretability</POST_TAG> technique, but has many subtleties regarding how it is applied and how one may interpret the results. We provide a summary of advice and best practices, based on our experience using this technique in practice. We include an overview of the different ways to apply <PRE_TAG>activation patching</POST_TAG> and a discussion on how to interpret the results. We focus on what evidence patching experiments provide about <PRE_TAG>circuits</POST_TAG>, and on the choice of <PRE_TAG>metric</POST_TAG> and associated pitfalls.