Universal and Transferable Attacks on Aligned Language Models (coup probes)

Train transformer language models with reinforcement learning.

Representation Engineering: A Top-Down Approach to AI Transparency

Simple and easily configurable grid world environments for reinforcement learning

PyTorch version of Stable Baselines, reliable implementations of reinforcement learning algorithms.

Library for Model Based RL

Multi Task RL Baselines

A collection of robotics simulation environments for reinforcement learning

Hierarchical Universal Language Conditioned Policies

CALVIN - A benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Tasks

Simple and easily configurable 3D FPS-game-like environments for reinforcement learning

An offline deep reinforcement learning library

Pybullet version of the multigoal robotics environment from OpenAI Gym

DrQ-v2: Improved Data-Augmented Reinforcement Learning

An open source reinforcement learning framework for training, evaluating, and deploying robust trading agents.

Proto-RL: Reinforcement Learning with Prototypical Representations

SenseAct: A computational framework for developing real-world robot learning tasks