align sampling param ability according to rfc

[aolemila]

RFC: sgl-project#202

[gemini-code-assist]

Summary of Changes

Hello @aolemila, 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 enhances the token generation and management system by introducing the capability to return raw token IDs alongside the decoded text. It also improves the robustness of the system by adding mechanisms to filter special tokens and handle invalid token IDs that fall outside the defined vocabulary, ensuring more compliant and predictable output generation, likely in accordance with an RFC.

Highlights

• Token ID Output: The system now returns generated token IDs alongside the decoded text, enabling more detailed analysis or downstream processing.
• Special Token Filtering: A new utility function has been added to filter out special tokens from the output token ID list based on a skip_special_tokens parameter.
• Invalid Token Handling: Robust error handling has been implemented to detect and manage out-of-vocabulary or invalid token IDs during generation, replacing them with stop tokens to prevent unexpected behavior.
• Vocab Size Integration: The vocabulary size is now passed to the request object, allowing for validation of generated token IDs.

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

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 the capability to return output token IDs alongside the decoded text and adds robustness to the sampling process. My review focuses on improving code quality, fixing a logic bug, and enhancing robustness. Specifically, I've suggested removing an unused import, refactoring a function for clarity and correctness, and fixing a bug in the stop token replacement logic. I also pointed out a potential TypeError and suggested improvements for determinism and logging clarity.