Skip to content

Fix kwarg vs pos arg hidden states #1694

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
merged 3 commits into from
Aug 20, 2025
Merged

Fix kwarg vs pos arg hidden states #1694

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
947d6ea
Update model_test.py
LRL2-ModelCloud Aug 20, 2025
4cdb84b
fix unclear hidden_states
LRL2-ModelCloud Aug 20, 2025
f576c6b
Merge branch 'main' into fix-unclear-inp
LRL2-ModelCloud Aug 20, 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
15 changes: 7 additions & 8 deletions gptqmodel/looper/module_looper.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -60,14 +60,13 @@ def cache_inputs(self, layers, auto_gc, calibration_data, calibration_enable_gpu
def store_input_hook(module, args, kwargs):
# Positional arguments.
layer_input = []
for inp in args:
layer_input.append(move_to(inp, device=data_device))
if len(layer_input) == 0:
# Some models put hidden_states in kwargs instead of args.
# For example, gptj ...
if kwargs.get("hidden_states") is not None:
layer_input.append(move_to(kwargs["hidden_states"], device=data_device))

if kwargs.get("hidden_states") is not None:
layer_input.append(move_to(kwargs["hidden_states"], device=data_device))
else:
# If hidden_states is not in kwargs, get it from the first positional argument
# If error occurs here, check the model's modeling code
layer_input.append(move_to(args[0], device=data_device))

layer_inputs.append(layer_input)

# Keyword arguments.
Expand Down