Models know when they're reward hacking
Article URL: https://www.goodfire.com/research/reward-hacking-activation-monitors
Comments URL: https://news.ycombinator.com/item?id=50116630
Points: 1
# Comments: 0
原文:https://www.a8q9.cn/en/article/models-know-when-they-re-reward-hacking-61f4b8.html