The Fine-Tuning Index / RLHF & Preference / #173

hhnqqq/MyTransformers

by hhnqqq · RLHF & Preference · updated 7mo ago

This repository provides a comprehensive library for parallel training and LoRA algorithm implementations, supporting multiple parallel strategies and a rich collection of LoRA variants. It serves as a flexible and efficient model fine-tuning toolkit for researchers and developers. Please contact hehn@mail.ustc.edu.cn for detailed information.

21
momentum
68
stars
9
forks
#173
rank
View on GitHub →