Florian Brand
@xeophon
evals @PrimeIntellect | open models @interconnectsai
加入 July 2015
843 正在关注 17.7K 粉丝
On the public TerminalBench traces, Astra runs into safety classifiers and their rollouts get stopped and scored as 0
Fable gets re-routed to Opus instead and the run continues.
Hard to compare scores that way!
显示更多
