Skip to content

[bugfix] fix gptq_bridge#19

Merged
Jintao-Huang merged 2 commits intomodelscope:mainfrom
Jintao-Huang:fix_gptq_bridge
Apr 8, 2026
Merged

[bugfix] fix gptq_bridge#19
Jintao-Huang merged 2 commits intomodelscope:mainfrom
Jintao-Huang:fix_gptq_bridge

Conversation

@Jintao-Huang
Copy link
Copy Markdown
Collaborator

No description provided.

@Jintao-Huang Jintao-Huang merged commit 4ef44f1 into modelscope:main Apr 8, 2026
1 check passed
Copy link
Copy Markdown

@gemini-code-assist gemini-code-assist bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request updates the _set_layer_mlp method in qwen3_next.py to include an is_mtp parameter, ensuring it is correctly propagated to the superclass and internal state-setting functions. A review comment suggests wrapping a long line in the updated code to improve readability.

# dense
mg_mlp = None if mg_layer is None else mg_layer.mlp
hf_state_dict.update(self._set_mlp_state(mg_mlp, hf_state_dict, f'{self.hf_mlp_prefix}.', layer_idx, to_mcore))
hf_state_dict.update(
Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

This line is quite long and exceeds typical readability limits. Consider wrapping the function call to improve clarity, especially since the subsequent _set_state_dict call is already wrapped.

        hf_state_dict.update(
            self._set_mlp_state(mg_mlp, hf_state_dict, f'{self.hf_mlp_prefix}.', layer_idx, to_mcore, is_mtp=is_mtp))

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant