i find it entertaining that under the hood, most open source "GRPO" implementations (eg. trl) by default actually implement REINFORCE with monte carlo group-advantages (by not reusing rollouts & making clipping/ratios redundant)

hello friends, i will be in SF for july/aug/some of sept - if you know any summer sublets/rentals in the city i should look at on short notice, dm me :)

discussing classic literature with frontier models really exposes their overconfidence. presumably because Dickens appears often in high-quality pretraining corpora, gpt-4o believes it can respond to all my questions without searching, resulting in near-constant hallucination.

blake was an unreasonably generous research mentor when i was a naive college sophomore, and it changed the course of my life. i thought back then that's what all grad students were like, but i soon realized he was uniquely prolific and polymathic. go work with him!