← All symbols

Symbol in context

mpatchm_{\text{patch}}

This entry links every published equation using this exact notation. Its meaning may change between equations.

Meanings in context

Used in 2 equations

score(C)=mpatch(C)−mcorruptmclean−mcorrupt,\text{score}(\mathcal C) = \frac{m_{\text{patch}}(\mathcal C) - m_{\text{corrupt}}}{m_{\text{clean}} - m_{\text{corrupt}}},

Probing, Sparse Autoencoders, Patching, and Steering: The Main Interpretability Methods, Compared · Equation 17

This equation states an equality: the expressions on both sides have the same value under the article’s assumptions.

Equation guide → · This term → · Article →