The study investigates LLMs using Fakepedia, a dataset presenting contradictions between known facts and new information. Through Masked Grouped Causal Tracing (MGCT), the research deciphers LLMs' grounding mechanisms by contrasting neural activation patterns. Findings help understand the co-functioning of grounding with recall capabilities within LLMs.
This page was last edited on 2024-05-03.
This page was last edited on 2024-05-03.