Skip to content

feat(moe): add gshard token rearrange optim #1127

feat(moe): add gshard token rearrange optim

feat(moe): add gshard token rearrange optim #1127

Annotations

2 errors and 1 warning

training_16GPU_4DP2TP2PP_MTP (910B)

succeeded Oct 25, 2024 in 7m 59s