Peter Hase (@peterbhase) 's Twitter Profile
Peter Hase

@peterbhase

AI safety researcher. PhD from UNC Chapel Hill (Google PhD Fellow). Previously: Anthropic, AI2, Google, Meta

ID: 1119252439050354688

linkhttps://peterbhase.github.io/ calendar_today19-04-2019 14:52:30

447 Tweet

3,3K Takipçi

960 Takip Edilen

Peter Hase (@peterbhase) 's Twitter Profile Photo
Peter Hase
@peterbhase

a month ago

Shower thought: LLMs still have very incoherent notions of evidence, and they update in strange ways when presented with information in-context that is relevant to their beliefs. I really wonder what will happen when LLM agents start doing interp on themselves and see the source