Continuous-Depth Field Theory for Transformer Patching and Mechanistic Interpretability 文章

ArXiv CS.AI2026-05-26NEWSen作者: David N. Olivieri, Antonio F. P\'erez Rodr\'iguez
大语言模型
查看原文 →
关系图谱

Continuous-Depth Field Theory for Transformer Patching and Mechanistic Interpretability · 相关技术

相关技术

AutoregressiveMechanistic InterpretabilityORM远程代码执行（RCE）residual streaminterpretabilityTokenTEEStraight-Through EstimatorSpatial Pivot-Aligned Coordinate-free Embedding (SPACE)SPAReferring expression comprehension (REC)MECICTHISGANForANNtransformersGitGPT