Skip to content

[mxfp8 training] add TP warning #2562

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

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

Open
danielvegamyhre wants to merge 1 commit into
base: main
Choose a base branch
from
+6 −0
Open

[mxfp8 training] add TP warning #2562

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions torchtitan/components/quantization/mx.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -84,6 +84,12 @@ def __init__(
"torch.compile enablement is required for highest performance of MXFP8 dynamic quantization."
)

if parallel_dims.tp > 1:
logger.warning(
"TP support for MXFP8 linears is still in progress. "
"Any linear layers with TP applied will use default precision, not MXFP8."
)

self.config = config
self.enabled = True
logger.info("MXFP8 MoE training enabled")
Expand Down
Loading