x
AI Safety Thursdays: Are LLMs aware of their learned behaviors? — LessWrong