Introducing MiniGPT

A PyTorch implementation of pre-training and fine-tuning scripts to train and fine-tune GPT-2 models

By Michael Hu
July 17, 2023 10:00 pm
2 min read

Transformer based language models like GPT are a hot topic these days, with the success of ChatGPT in particular. We are excited to introduce our most recent project MiniGPT, an open-source implementation of OpenAI's GPT-2 model [1].

The project was implemented in PyTorch and provides comprehensive support for dataset preparation, pre-training, fine-tuning, and distributed training with PyTorch FSDP.

References

  • [1]

    Radford, Alec, Wu, Jeff, Child, Rewon, Luan, David, Amodei, Dario, Sutskever, Ilya. Language Models are Unsupervised Multitask Learners. 2019.