conftrace_

interpretability research

3 papers

Explore in graph

Co-occurring keywords · papers mentioning both

language model (2) mechanism competition (1) logit inspection (1) attention modification (1) neural network (1) computation graph pruning (1) sheaf discovery (1) weight parameter (1) mechanism analysis (1) linear representation hypothesis (1)

Papers

Sheaf Discovery with Joint Computation Graph Pruning and Flexible Granularity EMNLP 2025 Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals ACL 2024 Recurrent Neural Networks Learn to Store and Generate Sequences using Non-Linear Representations EMNLP 2024