Add a blurb for DeepSeek v3 fine-tuning FP8 Recipe

[BirdsOfAFthr]

Description

Add a blurb for DeepSeek v3 fine-tuning FP8 Recipe.

Checklist

Before submitting this PR, please make sure (put X in square brackets):

• I have performed a self-review of my code. For an optional AI review, add the gemini-review label.
• I have necessary comments in my code, particularly in hard-to-understand areas.
• I have run end-to-end tests tests and provided workload links above if applicable.
• I have made or will make corresponding changes to the doc if needed, including adding new documentation pages to the relevant Table of Contents (toctree directive) as explained in our documentation.

[RissyRan]

Minor comments, LGTM overall! Thank you

[suexu1025]

Thanks @BirdsOfAFthr! Could you also briefly include the specific details of the recipe? Like clarify the FW/BW FP8 formats, the quantization granularity used here, and how this configuration was validated regarding convergence.

[github-actions]

🤖 Hi @RissyRan, I've received your request, and I'm working on it now! You can track my progress in the logs for more details.

[github-actions]

📋 Review Summary

This Pull Request introduces a new section documenting the DeepSeek V3 Fine-tuning FP8 Recipe. The content is well-structured and provides valuable information regarding the quantization scope, FP8 recipe, convergence, and performance sensitivity.

🔍 General Feedback

• The new documentation is clear and informative, enhancing the overall quality of the quantization.md file.

[RissyRan]

Thanks! 3 minor comments.