Confidence and Calibration of Activation Oracles for Reliable Interpretation of Language Model Internals

Add code
May 25, 2026

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: