fix draft config loading

[ggg-s]

Motivation

Root Cause
In the build_draft_model() function at line 350, when --ckpt-dir is provided, the code incorrectly assigns a string path to draft_model_config instead of loading the config object:

Line 350 (before fix)
draft_model_config = os.path.join(args.ckpt_dir, "config.json") # This is a string!
This causes the error later at line 414 when trying to access draft_model_config.vocab_size.
Solution
Replace line 350 with proper config loading:

Fixed code
draft_model_config = AutoDraftModelConfig.from_file(
os.path.join(args.ckpt_dir, "config.json")
)

Modifications

scripts/train_eagle3.py

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://sgl-fru7574.slack.com/archives/C09784E3EN6 to discuss your PR.

[gemini-code-assist]

Summary of Changes

Hello @ggg-s, 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 resolves a critical bug in the build_draft_model function that prevented the correct loading of draft model configurations when a checkpoint directory was specified. The change ensures that the configuration is properly parsed into an object, thereby enabling the training process to correctly utilize pre-trained draft models without encountering type-related errors.

Highlights

• Incorrect Draft Model Configuration Loading: Previously, when the --ckpt-dir argument was provided, the draft_model_config variable was incorrectly assigned a string path to config.json instead of loading the actual configuration object. This led to runtime errors when attempting to access attributes like vocab_size from the string.
• Proper Configuration Instantiation: The fix addresses this by replacing the direct string assignment with a call to AutoDraftModelConfig.from_file(). This ensures that draft_model_config is correctly instantiated as an AutoDraftModelConfig object, allowing proper access to its attributes and resolving the underlying error.

🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

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 addresses a critical bug in loading the draft model configuration. Previously, a file path string was assigned to draft_model_config instead of the configuration object itself, leading to an AttributeError. The fix correctly uses AutoDraftModelConfig.from_file to load the configuration from the JSON file, which is the proper way to handle this and resolves the issue. The change is correct and well-targeted.