Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

TAP: An automated jailbreaking method for black-box LLMs

Monitoring lab machines

A fast + lightweight implementation of the GCG algorithm in PyTorch

🤗 smolagents: a barebones library for agents that think in code.

Student version of Assignment 1 for Stanford CS336 - Language Modeling From Scratch

State-of-the-art TTS model under 25MB 😻

A CLI tool that helps AI researchers share datasets responsibly.

Code for the Paper: "STAMP Your Content: Proving Dataset Membership via Watermarked Rephrasings"

Code and training scripts for FlexOlmo

Retrieval and Retrieval-augmented LLMs

MTEB: Massive Text Embedding Benchmark

Data for the MTEB leaderboard

Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks [ICLR 2025]

Reference implementations for the MIDST challenge (Membership Inference over Diffusion-models-based Synthetic Tabular data) - SaTML 2025!

Code and data to go with the Zhu et al. paper "An Objective for Nuanced LLM Jailbreaks"

Train Models Contrastively in Pytorch

Implementation of the paper "MAZE: Data-Free Model Stealing Attack Using Zeroth-Order Gradient Estimation".