AI 学者图谱AI Scholar Graph ↗
Chris OlahChris Olah
Anthropic
神经网络可解释性先驱Pioneer of Neural Network Interpretability
safetydeep-learning
约1991年生于多伦多。毕业于 The Abelard School(2010年获 National AP Scholar)。18岁时从大学辍学,获得 Thiel Fellowship(Peter Thiel 设立、给年轻人 10 万美元跳过大学去创造的项目)。没有学位的他在 Google 做了一次演讲,Jeff Dean 印象深刻,给了他实习机会。在 Google Brain 实习2年,后共工作4年。因其在 colah.github.io 上发表的关于神经网络可视化的极其清晰的博客文章而闻名,被学术界广泛引用。后在 OpenAI 领导可解释性团队。2021年联合创办 Anthropic,现领导可解释性研究,在机械可解释性方面做出重要工作,绘制大语言模型内部的特征与电路图谱。AI 领域最知名的自学成才研究者之一。
Born around 1991 in Toronto. Graduated from The Abelard School (National AP Scholar, 2010). Dropped out of university at age 18 and received a Thiel Fellowship -- Peter Thiel's program paying young people $100K to skip college and build things. With no degree, gave a talk at Google that impressed Jeff Dean, who offered him an internship. Interned at Google Brain for 2 years, then stayed for 4 years total. Became known for his clear blog posts on neural network visualization (colah.github.io). Led the interpretability team at OpenAI. Co-founded Anthropic (2021) where he now leads interpretability research, producing work on mechanistic interpretability, mapping the internal features and circuits inside large language models. One of the most prominent self-taught researchers in AI.
同一人物 · 姊妹站:听 TA 的播客访谈(AI Podcast) · 读 TA 的论文与长文(AI Paper)
在关系图谱中查看 Chris Olah →View Chris Olah in the graph →
时间线Timeline
关系网络Connections
Dario Amodei联合创办 AnthropicJack Clark联合创办 Anthropic萨姆·麦坎德利什 Sam McCandlishAnthropic 联合创始人Boris ChernyAnthropic 同事尼尔·南达 Neel Nanda机械可解释性
← 返回完整图谱← Back to the graph