Arthur Conmy
@ArthurConmy
@anthropicai
prev fixing things @googledeepmind
๊ฐ์
August 2021
1.7K ํ๋ก์ ์ค 9.5K ํฌ
Our new @GoogleDeepMind paper studies novel activation probe architectures for classifying real-world misuse risks.
Our research has informed live deployments of probes in Gemini. ๐งต
๋ ๋ณด๊ธฐ
