clue-ai/ChatYuan

分布式训练是数据并行还是模型并行？感觉要模型并行才能成为大模型，是吧？

Closed this issue a year ago · 1 comments

guotong1988 commented 2 years ago

https://github.com/clue-ai/ChatYuan/tree/main/distributed-training

guotong1988 commented a year ago

https://github.com/NVIDIA/Megatron-LM