add kv quant example

[mengniwang95]

SUMMARY:
Add kv cache quantization into autoround example

TEST PLAN:
Have test the example offline

[gemini-code-assist]

Summary of Changes

Hello @mengniwang95, 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 integrates Key-Value (KV) cache quantization into the existing AutoRound example, allowing users to optionally apply FP8 quantization to the KV cache. This enhancement provides a mechanism to explore the impact of KV cache quantization on model performance and resource utilization, with updated documentation to guide users through the process and present comparative accuracy metrics.

Highlights

• KV Cache Quantization: Introduced the ability to quantize the Key-Value cache to FP8 within the AutoRound example, offering more optimization options for large language models.
• Conditional Quantization Logic: The example script (llama3_example.py) now conditionally applies KV cache quantization based on a new command-line argument, --fp8_kv, allowing users to easily toggle this feature.
• Documentation and Results Update: The README.md has been updated with clear instructions on how to enable KV cache quantization and includes comparative accuracy results, demonstrating the impact of this optimization.

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 adds an example for KV cache quantization with AutoRound. The changes include updating the llama3_example.py script to support a --fp8_kv flag and modifying the corresponding README.md to document this new feature, including usage instructions and evaluation results. My review focuses on improving the clarity and correctness of the documentation in the README.md file. I've pointed out a confusing comment, a typo, and an incorrect markdown code block specifier. The code changes in the Python script look good and are clear for an example.

[github-actions]

👋 Hi! Thank you for contributing to llm-compressor. Please add the ready label when the PR is ready for review.

Note: This is required to complete the testing suite, please only add the label once the PR is code complete and local testing has been performed.

[hshen14]

@dsikka @kylesayrs @brian-dellabetta please help review the PR. Thanks.