Skip to content

feat(model): add generic chunked linear cross entropy #177

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
maomaocun wants to merge 1 commit into
base: main
Choose a base branch
from
Open

feat(model): add generic chunked linear cross entropy #177

Show file tree
Hide file tree
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
5 changes: 5 additions & 0 deletions src/mcore_bridge/config/model_config.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -269,6 +269,11 @@ class ModelConfig(TransformerConfig):
num_labels: Optional[int] = None
mlp_padding_free: bool = False

# Optional memory-bounded LM-head + cross-entropy path. A value of zero
# keeps the regular output-layer path; the corresponding environment
# variable is useful for launchers that cannot extend ModelConfig.
chunked_linear_ce_chunk_size: int = 0

_mindspeed_defaults_cache = None

def _augment_mindspeed_defaults(self):
Expand Down
Loading