https://arxiv.org/abs/2410.19750 The Geometry of Concepts: Sparse Autoencoder Feature StructureSparse autoencoders have recently produced dictionaries of high-dimensional vectors corresponding to the universe of concepts represented by large language models. We find that this concept universe has interesting structure at three levels: 1) The "atomicarxiv.org원자 수준: 단어 관계(예: "Austria:Vienna::Switz..