Stanford NLP Python Library for Understanding and Improving PyTorch Models via Interventions

For OpenMOSS Mechanistic Interpretability Team's Sparse Autoencoder (SAE) research.

Mechanistically interpretable neurosymbolic AI (Nature Comput Sci 2024): losslessly compressing NNs to computer code and discovering new algorithms which generalize out-of-distribution and outperform human-designed algorithms

Steering vectors for transformer language models in Pytorch / Huggingface

Sparse and discrete interpretability tool for neural networks

CausalGym: Benchmarking causal interpretability methods on linguistic tasks

PyTorch and NNsight implementation of AtP* (Kramar et al 2024, DeepMind)

Multi-Layer Sparse Autoencoders (ICLR 2025)

graphpatch is a library for activation patching on PyTorch neural network models.

Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals

[ACL'2024 Findings] "Understanding and Patching Compositional Reasoning in LLMs"

A small package implementing some useful wrapping around nnsight

This repository contains the code used for the experiments in the paper "Discovering Variable Binding Circuitry with Desiderata".

MechaMap - Toolkit for Mechanistic Interpretability (MI) Research

Exploring length generalization in the context of indirect object identification (IOI) task for mechanistic interpretability.

A framework for conducting interpretability research and for developing an LLM from a synthetic dataset.

Starting Kit for the CodaBench competition on Transformer Interpretability