Repo for Paper: Discovering Interpretable Algorithms by Decompiling Transformers to RASP

A special type of transformer with discrete embeddings and attention.

2026 AI/ML internship & new graduate job list updated daily

Dafny is a verification-aware programming language

Train the smallest LM you can that fits in 16MB. Best model wins!

Deep Learning Interpretability with Symbolic Regression

Open-source release accompanying Gao et al. 2025

Open-source implementation of AlphaEvolve

egg is a flexible, high-performance e-graph library

An interpreter for RASP as described in the ICML 2021 paper "Thinking Like Transformers"

Python functions powered by AI agents - with runtime post-conditions for reliable agentic workflows.

High-Performance Symbolic Regression in Python and Julia

Distributed High-Performance Symbolic Regression in Julia

OpenAI & Ollama compatible API powered by Codex

Google's Operations Research tools:

The math library of Lean 4

A Foundation Model for Generalist Gaming Agents

A Lean companion to Analysis I

The Unity Machine Learning Agents Toolkit (ML-Agents) is an open-source project that enables games and simulations to serve as environments for training intelligent agents using deep reinforcement …

Lean 4 programming language and theorem prover

This repository contains implementations and illustrative code to accompany DeepMind publications

Open source code for AlphaFold 2.

AlphaFold 3 inference pipeline.

Project repo for gpSLDS

Recurrent Switching Linear Dynamical Systems

Code supporting the paper "Weakly Supervised Learning with Assemblies of Neurons", by Dabagia, Papadimitriou, and Vempala [2021].

Hackable and optimized Transformers building blocks, supporting a composable construction.