Link model to paper and improve model card
#1
by nielsr HF Staff - opened
Hi! I'm Niels from the Hugging Face community team.
This PR improves your model card by linking it to the corresponding paper page on Hugging Face: Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm.
I've also added the author list and provided more context about the SPES framework (SParse Expert Synchronization) to help users understand the decentralized training approach used for this 9B MoE model.
zjr2000 changed pull request status to merged