Dictionary Learning Improves Patch-Free Circuit Discovery in Mechanistic Interpretability: A Case Study on Othello-GPT
arXiv (Cornell University)(2024)
关键词
Inductive Modeling,Robust Learning,Function Approximation
AI 理解论文
溯源树
样例

生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要