Ethan Perez

website Personal homepage

core Anthropic

Alignment and RLHF Constitutional AI: Harmlessness from AI Feedback Alignment and Safety Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training Alignment and Safety Alignment faking in large language models Alignment and Safety Auditing language models for hidden objectives Alignment and Safety Constitutional Classifiers: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming Alignment and Safety Constitutional Classifiers++: Defending against Universal Jailbreaks across Thousands of Hours of Red Teaming Interpretability On the Biology of a Large Language Model Interpretability Tracing the thoughts of a large language model