bigscience-workshop / Megatron-DeepSpeed

Ongoing research training transformer language models at scale, including: BERT & GPT-2
Other
1.3k stars 211 forks source link

upgrade megatron-lm #378

Open dz1iang opened 1 year ago

dz1iang commented 1 year ago

HI,Is there plan to upgrade megatron-lm,it can be support more feature and better performance