Code accompanying the paper "Massive Activations in Large Language Models"

A library for mechanistic interpretability of GPT-style language models

Codes for our paper "Full-Atom Peptide Design with Geometric Latent Diffusion" (NeurIPS 2024)

[IEEE CISS 2024, ICMLW 2023] Assessing Neural Network Representations During Training Using Noise-Resilient Diffusion Spectral Entropy

Reference implementation for Deep Unsupervised Learning using Nonequilibrium Thermodynamics

Wraps PyTorch code in a JIT-compatible way for JAX. Supports automatically defining gradients for reverse-mode AutoDiff.

Run PyTorch in JAX. 🤝

Send data to and from pymol from a remote server (e.g. a cluster running deep learning workflows)

Protein Ligand INteraction Dataset and Evaluation Resource

Tasks Assessing Protein Embeddings (TAPE), a set of five biologically relevant semi-supervised learning tasks spread across different domains of protein biology.

Lagrangian formulation of Doob's h-transform allowing for efficient rare event sampling

EquiTriton is a project that seeks to implement high-performance kernels for commonly used building blocks in equivariant neural networks, enabling compute efficient training and inference.

Force Fields

User friendly and accurate binder design pipeline

Exploring the conformational ensembles of protein-protein complexes with transformer-based generative neural networks

Generative modeling of molecular dynamics trajectories

Protein-Nucleic Acid Complex Modeling with Frame Averaging Transformer, NeurIPS2024

Uncover meaningful structures of latent spaces learned by generative models with flows!

Efficient Triton Kernels for LLM Training

Adapting protein language models for structure-conditioned design

A Python package for processing molecules with RDKit in scikit-learn

Simple Scalable Discrete Diffusion for text in PyTorch

Metrics for assessing 3D molecular structures.

Foster the development of impactful AI models in drug discovery.

A native PyTorch Library for large model training