LLMs Explain't: A Post-Mortem on Semantic Interpretability in Transformer Models.

Alhassan Abdelhalim, Janick Edinger, Sren Laue, Michaela Regneri

Browse the full PERCOM paper archive.