Log in
Kye Gomez

Kye Gomez/BitNet

Implementation of "BitNet: Scaling 1-bit Transformers for Large Language Models" in pytorch

Issues looking for funding

I wonder what hardware conditions (GPU) the code uses, and why the loss value has been above 5.2 after running the train.py file, and the validation generated unreadable incomprehensible content.
bug

#68 opened in kyegomez/BitNet

Fund
Kye Gomez
kyegomez/BitNet
Creatorkyegomez
Stars1.7k
LicenseMIT License
RepositoryGitHub
Websitediscord.gg