Skip to content

louiezzang/next-gpt

Repository files navigation

nextGPT

📢 Open source implementation for ChatGPT replica to build the end-to-end pipeline from SFT to RLHF.

  • 🔥 Step 1) SFT: Surpervised Fine-tuning
  • 🔥 Step 2) RM: Reward Model
  • 🔥 Step 3) PPO: Proximal Policy Optimization

ChatGPT Diagram

Installation

$ pip install nextgpt

or install from the git repo to get always the latest version.

$ git clone https://github.com/louiezzang/next-gpt.git
$ cd next-gpt/
$ pip install .
$ cd ../

Examples

See chatGPT example

RLHF

What is RLHF?

Implementation of RLHF (Reinforcement Learning with Human Feedback) was powered by Colossal-AI. More details can be found in the blog.

The RLHF was forked and modified from these git repos.

References

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published