[FIX] vLLM LoRA extra vocab

[Datta0]

Also sync along with unslothai/unsloth#3662

[gemini-code-assist]

Summary of Changes

Hello @Datta0, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request focuses on enhancing the robustness and compatibility of the system, particularly concerning vLLM and LoRA integrations. It introduces a more flexible approach to loading LoRA adapters, ensuring proper handling of extended vocabularies across different API versions. Additionally, it refines the parameter handling for sampling functions and cleans up an obsolete patch related to vLLM's sampling parameters, streamlining the codebase.

Highlights

• LoRA Adapter Loading Improvement: Refactored the LoRA adapter loading mechanism to dynamically handle different API signatures for vocabulary size and embedding padding, ensuring better compatibility and correct application of lora_extra_vocab_size.
• Sampling Parameter Filtering: Implemented filtering for generation_kwargs in grpo_update_SamplingParams to ensure only valid sampling parameters are passed, preventing potential errors from unrecognized arguments.
• VLLM Patch Removal: Removed an outdated temporary patch that mapped GuidedDecodingParams to StructuredOutputsParams in vLLM, indicating that the underlying dependency issue has been resolved.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces a fix for vLLM LoRA extra vocabulary handling by adding backward compatibility for a parameter name change in vllm_lora_worker_manager.py. The change is well-implemented and improves code maintainability by reducing duplication. Additionally, the PR includes several stylistic cleanups by removing trailing whitespace and also removes a temporary patch from vllm_utils.py, which is a welcome cleanup. I have one suggestion in rl_replacements.py to make a code block more concise and Pythonic.

[gemini-code-assist]

This loop for filtering generation_kwargs can be made more concise and Pythonic by using a dictionary comprehension. This improves readability and is generally more performant.

Suggested change

	new_generation_kwargs = {}
	for key in generation_kwargs.keys():
	if key in good_sampling_params_keys:
	new_generation_kwargs[key] = generation_kwargs[key]
	generation_kwargs = new_generation_kwargs
	generation_kwargs = {k: v for k, v in generation_kwargs.items() if k in good_sampling_params_keys}