X · @teortaxesTex
· X / Twitter
I think in this case it's just that RL is currently easier logistically than pretraining on Ascends. I suspect it's just not a very good pretrain, kno…
I think in this case it's just that RL is currently easier logistically than pretraining on Ascends. I suspect it's just not a very good pretrain, knowledge-wise. Eventually most evals will *become* multi-step and agentic, but I expect V4 to report more normal stuff.Alexander Doria: Clearer mark of a new era: latest mo