Skip to content

Latest commit

 

History

History
6 lines (3 loc) · 256 Bytes

File metadata and controls

6 lines (3 loc) · 256 Bytes

This repository contains the official project page for the paper:

POET-X: Memory-efficient LLM Training by Scaling Orthogonal Transformation (paper).

The implementation of POET-X is host at https://github.com/Sphere-AI-Lab/poet