fix save checkpoint bug

[sleepcoo]

Motivation

Modifications

Related Issues

Accuracy Test

Benchmark & Profiling

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 @sleepcoo, 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 introduces a crucial fix to the training script scripts/train_eagle3.py by guaranteeing that a final checkpoint is always saved when training completes. This prevents potential loss of progress if the training process finishes at a step that does not align with the predefined save interval, thereby enhancing the robustness and reliability of the training workflow.

Highlights

• Checkpoint Saving Logic: Implemented a new conditional block to ensure a final checkpoint is saved at the end of the training process in scripts/train_eagle3.py. This addresses a potential bug where the last checkpoint might not be saved if training concludes between scheduled save intervals.

🧠 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 correctly identifies and fixes a bug where the final checkpoint might not be saved if training concludes on a step that is not a multiple of the save interval. The added logic ensures the final model state is preserved. I've included one suggestion to improve the robustness of the script by handling a potential edge case.

[gemini-code-assist]

To prevent a potential ZeroDivisionError if args.save_interval is set to 0, it's good practice to add a check to ensure args.save_interval is positive before performing the modulo operation. This would make the script more robust against invalid inputs. A similar check should also be considered for the checkpoint logic inside the training loop on line 855 for consistency.

Suggested change

	if global_step % args.save_interval != 0:
	print_on_rank0(
	f"Training completed at step {global_step}, saving final checkpoint..."
	)
	save_checkpoints(args, epoch, global_step, eagle3_model, optimizer)
	if args.save_interval > 0 and global_step % args.save_interval != 0:
	print_on_rank0(
	f"Training completed at step {global_step}, saving final checkpoint..."
	)
	save_checkpoints(args, epoch, global_step, eagle3_model, optimizer)