Depends on the training? If there was eg RLHF then those connections are stronge... | Hacker News

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

		pbhjpbhj 5 months ago \| parent \| context \| favorite \| on: My 2.5 year old laptop can write Space Invaders in... Depends on the training? If there was eg RLHF then those connections are stronger and more likely; that's a difference (but not a category difference).

0x457 4 months ago [–]

Yes, but I thought we're talking about category difference.

Proper RLHF surely boosts "predicted next token until it couldn't" to feel more like "actually recalled".

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