Update moe.sh

[Itime-ren]

PR type

• Bug Fix
• New Feature
• Document Updates
• More Models or Datasets Support

PR information

Write the detail information belongs to this PR.

Experiment results

Paste your experiment result here(if needed).

[gemini-code-assist]

Summary of Changes

Hello @Itime-ren, 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 updates the moe.sh script to optimize the configuration for a Megatron SFT (Supervised Fine-Tuning) job. The changes involve expanding the GPU resources used, redirecting the training dataset to a local file, and adjusting the frequency of evaluation and model saving. Additionally, it introduces integration with Weights & Biases for enhanced experiment tracking.

Highlights

• GPU Resource Allocation: The script now utilizes 4 CUDA devices (0,1,2,3) instead of 2, allowing for potentially faster or larger-scale training.
• Dataset Source Update: The dataset path has been changed from a Swift-specific reference to a local file system path: /root/autodl-tmp/swift_finetune_data/qwen3_finetune_self_cognition.jsonl.
• Training Interval Adjustments: The eval_interval and save_interval parameters have been significantly reduced from 200 to 5, leading to more frequent evaluation and checkpointing during training.
• Wandb Integration: A new argument --report_to wandb has been added, enabling integration with Weights & Biases for experiment tracking and reporting.

🧠 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 updates the moe.sh script, adjusting GPU settings, dataset paths, and training intervals. My review identifies a few key issues: a configuration mismatch between the number of visible GPUs and the number of processes, which will lead to underutilization of resources; the use of a hardcoded absolute path for the dataset, which harms the script's portability; and setting evaluation/saving intervals to a very low value, which may be suitable for debugging but is inefficient for general use. Addressing these points will improve the script's correctness and reusability.

[gemini-code-assist]

There's an inconsistency between NPROC_PER_NODE (set to 2 on the preceding line) and CUDA_VISIBLE_DEVICES (set to use 4 GPUs). This configuration will only utilize 2 of the 4 specified GPUs. To make use of all available GPUs, NPROC_PER_NODE should be updated to 4.

[gemini-code-assist]

The dataset is specified with a hardcoded absolute path (/root/...). This makes the script non-portable and dependent on a specific user's environment, which is not ideal for an example script in a shared repository. Please consider using a relative path, an environment variable, or reverting to the previous method of using dataset identifiers from a registry to ensure the script is reusable.

[gemini-code-assist]

The eval_interval and save_interval have been set to a very low value of 5. While this can be useful for rapid debugging, it will cause frequent evaluations and checkpointing, introducing significant overhead and consuming disk space quickly during a normal training run. If this change is for debugging, it should not be merged. For a general-purpose example script, a more conservative value (like the previous 200) is recommended.