We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Your question
pp
context_parallel_size > 1
encoder_tensor_parallel_size != tensor_parallel_size
When enabling context_parallel, the input will be split symmetrically to balance the calculation. Using zip(cycle(e_ranks), d_ranks) is wrong.
context_parallel
zip(cycle(e_ranks), d_ranks)
Megatron-LM/megatron/core/parallel_state.py
Line 602 in 4ff9e66
stack
Megatron-LM/megatron/core/pipeline_parallel/p2p_communication.py
Line 402 in 46ca068
The text was updated successfully, but these errors were encountered:
No branches or pull requests
Your question
pp
groups when enablingcontext_parallel_size > 1
andencoder_tensor_parallel_size != tensor_parallel_size
.When enabling
context_parallel
, the input will be split symmetrically to balance the calculation. Usingzip(cycle(e_ranks), d_ranks)
is wrong.Megatron-LM/megatron/core/parallel_state.py
Line 602 in 4ff9e66
stack
operator to calculate the sum of received tensor.Megatron-LM/megatron/core/pipeline_parallel/p2p_communication.py
Line 402 in 46ca068
The text was updated successfully, but these errors were encountered: