Skip to content
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

Fix saving adapter weights after disabling DSD #2351

Merged
merged 2 commits into from
Feb 6, 2025
Merged

Fix saving adapter weights after disabling DSD #2351

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f97896b
fix
jessicazhongeee Feb 6, 2025
8e8acff
misc
jessicazhongeee Feb 6, 2025
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
3 changes: 3 additions & 0 deletions torchtune/training/_distributed.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,7 @@
from torchtune.modules.attention import MultiHeadAttention
from torchtune.modules.model_fusion import DeepFusionModel

from torchtune.modules.peft import get_adapter_state_dict
from torchtune.utils import get_device, get_logger
from torchtune.utils._logging import deprecated

Expand Down Expand Up @@ -380,6 +381,8 @@ def gather_cpu_state_dict(
if is_rank_zero:
cpu_state_dict[param_name] = param.cpu()
torch.distributed.barrier()
if adapter_weights_only:
cpu_state_dict = get_adapter_state_dict(cpu_state_dict, device=None)
return cpu_state_dict


Expand Down