conftrace_

neural network circuit

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

mechanistic interpretability (2) subnetwork composition (1) representation theory (1) universality hypothesis (1) model weight analysis (1) neural network (1) modular structure (1) language model (1) group theory (1)

Papers

Circuit Compositions: Exploring Modular Structures in Transformer-Based Language Models ACL 2025 A Toy Model of Universality: Reverse Engineering how Networks Learn Group Operations ICML 2023