vwxyzjn

cleanrl

High-quality single file implementation of Deep Reinforcement Learning algorithms with research-friendly features (PPO, DQN, C51, DDPG, TD3, SAC, PPG)

ppo-implementation-details

The source code for the blog post The 37 Implementation Details of Proximal Policy Optimization

portwarden

Create Encrypted Backups of Your Bitwarden Vault with Attachments

lm-human-preference-details

RLHF implementation details of OAI's 2019 codebase

invalid-action-masking

Source Code for A Closer Look at Invalid Action Masking in Policy Gradient Algorithms

cleanba

CleanRL's implementation of DeepMind's Podracer Sebulba Architecture for Distributed DRL

summarize_from_feedback_details

benchmark-ci

free-mujoco-py

MuJoCo is a physics engine for detailed, efficient rigid body simulations with contacts. mujoco-py allows using MuJoCo from Python 3.

lm-human-preferences

Code for the paper Fine-Tuning Language Models from Human Preferences

minimal-adam-difference

ppo-atari-metrics

microrts

trl

Train transformer language models with reinforcement learning.

alignment-handbook

Robust recipes for to align language models with human and AI preferences

direct-preference-optimization

Reference implementation for DPO (Direct Preference Optimization)

cleanba-test

envpool_bug

hfblog

Public repo for HF blog posts

optax

Optax is a gradient processing and optimization library for JAX.

quickchat

transformers

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

zero3_min_repro

2024

MOSS-RLHF

MOSS-RLHF

peft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

PokemonRedExperiments

Playing Pokemon Red with Reinforcement Learning

summarize-from-feedback

Code for "Learning to summarize from human feedback"

tyro

Strongly typed, zero-effort CLI interfaces & config objects

vllm

A high-throughput and memory-efficient inference and serving engine for LLMs