Mechanistic Interpretability Visualizations using React

Code to reproduce key results accompanying "SAEs (usually) Transfer Between Base and Chat Models"

A curation of awesome tools, documents and projects about LLM Security.

A framework for few-shot evaluation of language models.

Representation Engineering: A Top-Down Approach to AI Transparency

Code and results accompanying the paper "Refusal in Language Models Is Mediated by a Single Direction".

Sparse Autoencoder for Mechanistic Interpretability

Training Sparse Autoencoders on Language Models

Sparse Autoencoder for Mechanistic Interpretability

Training Sparse Autoencoders on Language Models

LLM experiments done during SERI MATS - focusing on activation steering / interpreting activation spaces

This repository contains the code used for the experiments in the paper "Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking".

A scientific instrument for investigating latent spaces

Code for internal lab sharing - polishing has started but is by no means complete

A beautiful, simple, clean, and responsive Jekyll theme for academics

Hopfield Networks is All You Need

Deep Learning - Spring 2022 - Final Project