Thread

Fixing GRPO's credit assignment problem without evaluating every step

19 points · 3 comments · mrkn1

Loading the complete thread in the background. This saved snapshot is available now. Refresh

vatsachak · 1790963116 · focus · HN ↗
How do frontier labs choose which ideas to include in a training run? It seems that there are too many to choose from
1. martianvoid · 1790963758 · focus · HN ↗
  I think maybe they have enough compute to try out all the promising directions at a smaller scale first and then port them to larger run
  They also have incredibly smart people that can understand which improvements and directions are even worth pursuing for, I think these people are being paid in millions
2. simianwords · 1790965052 · focus · HN ↗
  Why do you think they are buying compute and accelerating to RSI? It’s for this.

Unofficial Hacker News client; not affiliated with Y Combinator.