Google DeepMind 机制可解释性负责人 · Mechanistic Interpretability Lead, Google DeepMind
领导 Google DeepMind 的机制可解释性研究——逆向工程大模型内部到底发生了什么。以开源可解释性工具与大量培养新人著称。
Leads mechanistic interpretability research at Google DeepMind — reverse-engineering what actually happens inside large models. Known for open-source interpretability tooling and prolific mentoring.
在 AI Podcast 查看 TA 的全部内容 →