search

NVIDIA / Megatron-LM

Ongoing research training transformer models at scale
https://docs.nvidia.com/megatron-core/developer-guide/latest/user-guide/index.html#quick-start
Other
9.23k stars 2.08k forks source link

[ENHANCEMENT]Can we pass a tuple that includs all the tensors I'd like to pass between pipeline's different stages? #867

Open janelu9 opened 2 weeks ago

janelu9 commented 2 weeks ago

It seems I can only pass one output hidden tesnor now