Hacker News new | past | comments | ask | show | jobs | submit
r_lee | on: Show HN: I RL-trained an agent that trains models with RL (for ~$1.3k)
why...?
snerbles | parent | next
I guess we found the target audience for models that slather READMEs in emojis.
loading story #48908804
piratebroadcast | parent
To circumvent distillation.