Skip to content

FIX low_cpu_mem_usage=True with 8bit bitsandbytes #2325

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

Merged
BenjaminBossan merged 2 commits into from
Jan 14, 2025
Merged

FIX low_cpu_mem_usage=True with 8bit bitsandbytes #2325

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
9174b90
FIX low_cpu_mem_usage=True with 8bit bitsandbytes
BenjaminBossan Jan 13, 2025
7dc36f7
Better fix: remove all the code
BenjaminBossan Jan 13, 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
Prev Previous commit
Better fix: remove all the code
  • Loading branch information
@BenjaminBossan
BenjaminBossan committed Jan 13, 2025
commit 7dc36f71c9c56e179c186626f5fffec1f2030a03
9 changes: 0 additions & 9 deletions src/peft/tuners/lora/model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -248,15 +248,6 @@ def _replace_module(self, parent, child_name, new_module, child):
child = child.base_layer

meta = torch.device("meta")

# handle .state attribute of 8 bit bnb
if (getattr(child, "state", None) is not None) and not any(p.device == meta for p in new_module.parameters()):
if hasattr(new_module, "base_layer"):
new_module.base_layer.state = child.state
else:
new_module.state = child.state
new_module.to(child.weight.device)

# dispatch to correct device
for name, module in new_module.named_modules():
if (self.prefix in name) or ("ranknum" in name):
Expand Down
Loading