AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
4min chapter
Collin Burns On Discovering Latent Knowledge In Language Models Without Supervision
The Inside View
I've Been Like Oh Yeah Ask Him if I Have a Sandwich and Not Ask Him
i think there's actually a lot of structure here in many cases like with if you have a lying model it should have some features that are correlated with when it's lying or something like that. i think the thing i'm excited about is something like honesty so i think truthfulness suggests that a model always outputs the truth. I don't feel very committed philosophically to any definition of police and stuff like i don't want to get the philosophers meant don't don't tell me right. In some sense the best we can hope for is getting models beliefs and also beliefs and sort of associate beliefs being associated with honesty rather than truthfulness. If we did have this superhuman model i would
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode