Skip to content
conftrace_

interchange intervention training

2 papers

Explore in graph

Also known as

IIT

Co-occurring keywords · papers mentioning both

neural network (2) causal abstraction (1) counterfactual behavior (1) circuit discovery (1) causal model (1) mechanistic interpretability (1) neural network verification (1) causal structure (1) transformer architecture (1) causal inference (1)

Papers

InterpBench: Semi-Synthetic Transformers for Evaluating Mechanistic Interpretability Techniques NeurIPS 2024 Inducing Causal Structure for Interpretable Neural Networks ICML 2022