2026-10-15 · America/Los_Angeles · 社区动态 · #3

interpretability might be the most important scientific problem in the world right now, as I don't think we can solve…

interpretability might be the most important scientific problem in the world right now, as I don't think we can solve alignment if we don't know what we're building. I think I have good news about it. I co-authored early work in mechanistic interpretability at OpenAI and came

热度 67.4 / 100；排名与评分保留该期记录。