Noob question - how do we know that these autoencoders aren't hallucinating and ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		wg0 on Dec 23, 2024 \| parent \| context \| favorite \| on: Show HN: Llama 3.3 70B Sparse Autoencoders with AP... Noob question - how do we know that these autoencoders aren't hallucinating and really are mapping/clustering what they should be?

trq_ on Dec 24, 2024 [–]

Hmm the hallucination would happen in the auto labelling, but we review and test our labels and they seem correct!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact