⚡ From the authors of Microsoft DeepSpeed - DeepSpeed-Chat! A new library that allows you to easily train 10B+ ChatGPT-style models on a single GPU! Single script, RLHF support, and 15x faster than SOTA apparently! GitHub: https://t.co/2oe2F0OFeu 1/ 🧵👇 https://t.co/poyD49QOfo
— Aleksa Gordić 🍿🤖 (@gordic_aleksa) Apr 14, 2023
from Twitter https://twitter.com/gordic_aleksa
April 14, 2023 at 07:44AM
via IFTTT
No comments:
Post a Comment