Skip to content

Eric Wallace: Memorization in language models

About this video

Hand-picked by a BrighterTube curator · found via round-3 back catalogue of a channel Lumen picks video by video

Memorization mechanisms in language models are analyzed by Wallace with concrete examples. A 66-minute lecture teaches how to identify and study unwanted model behaviors.

Video by UC Berkeley EECS, played through YouTube's embedded player.

Part of Understanding Large Language Models: Foundations and Safety