ndif-team/nnterp

A library for mechanistic interpretability of GPT-style language models

LLM Transparency Tool (LLM-TT), an open-source interactive toolkit for analyzing internal workings of Transformer-based language models. *Check out demo at* https://huggingface.co/spaces/facebook/llm-transparency-tool-demo

Delphi was the home of a temple to Phoebus Apollo, which famously had the inscription, 'Know Thyself.' This library lets language models know themselves through automated interpretability.

Companion code for the global workspace interpretability paper

Training Sparse Autoencoders on Language Models

🪄 Interpreto is an interpretability toolbox for LLMs

Layer-wise Relevance Propagation for Large Language Models and Vision Transformers [ICML 2024]

Interpretability for sequence generation models 🐛 🔍