You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
code: the main branch https://github.com/NVIDIA/Megatron-LM
transformer_engine 1.13
eager is normal, compile will report The provided qkv memory layout is not supported!
The error is as follows:
According to the functional description, the stride of k and v should be (SHD, HD, D, 1), but it is actually (HD, H*D, D, 1)
megatron/core/extensions/transformer_engine.py
printed result
Then call DotProductAttention, v'stride will become (524288, 128, 32, 1) directly,k'stride not change.
printed result
The text was updated successfully, but these errors were encountered:
code: the main branch https://github.com/NVIDIA/Megatron-LM
transformer_engine 1.13
eager is normal, compile will report The provided qkv memory layout is not supported!
The error is as follows:
According to the functional description, the stride of k and v should be (SHD, HD, D, 1), but it is actually (HD, H*D, D, 1)
megatron/core/extensions/transformer_engine.py
printed result
Then call DotProductAttention, v'stride will become (524288, 128, 32, 1) directly,k'stride not change.
printed result
The text was updated successfully, but these errors were encountered: