don't create a reference model if grpo beta is 0.0

[winglian]

Description

see https://github.com/huggingface/trl/blob/23ea671c5e7c83031c617171741ceabbf3fe18cc/trl/trainer/grpo_trainer.py#L726-L728

We can save memory here if not doing kl-divergence

Motivation and Context

How has this been tested?

Screenshots (if appropriate)

Types of changes

Social Handles (Optional)

Summary by CodeRabbit

• Bug Fixes
  • Improved handling of reference model loading during reinforcement learning training, ensuring the reference model is not loaded when using GRPO RL type with a beta value of zero.

[coderabbitai]

📝 Walkthrough

Walkthrough

The update refines the logic within the setup_reference_model function in src/axolotl/train.py. It introduces a conditional check to determine whether to load a reference model based on the RL type and the trl.beta parameter, specifically skipping the reference model when using GRPO with trl.beta set to zero.

Changes

File(s)	Change Summary
src/axolotl/train.py	Adjusted setup_reference_model to conditionally load the reference model based on RL type and parameters.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~7 minutes

---

📜 Recent review details

Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between f7ea140 and 7347550.

📒 Files selected for processing (1)

• src/axolotl/train.py (1 hunks)

🧰 Additional context used

🧠 Learnings (2)

📓 Common learnings

Learnt from: NanoCode012
PR: axolotl-ai-cloud/axolotl#2952
File: src/axolotl/common/datasets.py:125-125
Timestamp: 2025-07-22T08:27:00.129Z
Learning: GRPO (cfg.rl == "grpo" or cfg.rl is RLType.GRPO) should be excluded from dataset label checking during preprocessing, while other RL methods should continue to have this functionality available.

src/axolotl/train.py (1)

Learnt from: NanoCode012
PR: #2952
File: src/axolotl/common/datasets.py:125-125
Timestamp: 2025-07-22T08:27:00.129Z
Learning: GRPO (cfg.rl == "grpo" or cfg.rl is RLType.GRPO) should be excluded from dataset label checking during preprocessing, while other RL methods should continue to have this functionality available.

🧬 Code Graph Analysis (1)

src/axolotl/train.py (2)

src/axolotl/utils/schemas/enums.py (1)

• RLType (24-32)

src/axolotl/loaders/model.py (1)

• ModelLoader (63-805)

⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (8)

• GitHub Check: PyTest from Source Dist (3.11, 2.7.1)
• GitHub Check: PyTest from Source Dist (3.11, 2.7.0)
• GitHub Check: PyTest from Source Dist (3.11, 2.6.0)
• GitHub Check: PyTest (3.11, 2.7.1)
• GitHub Check: PyTest (3.11, 2.6.0)
• GitHub Check: PyTest (3.11, 2.7.0)
• GitHub Check: pre-commit
• GitHub Check: pre-commit

🔇 Additional comments (2)

src/axolotl/train.py (2)

118-120: LGTM: Memory optimization correctly implemented.

The conditional logic properly implements the memory optimization by skipping reference model creation when GRPO is used with beta=0. This aligns with the TRL implementation pattern and the retrieved learning that GRPO requires special handling.

---

122-122: LGTM: Correct parameter usage for conditional reference model loading.

The ModelLoader is now properly instantiated with the conditionally determined reference_model parameter, enabling the memory optimization when GRPO beta is 0.

✨ Finishing Touches

• 📝 Generate Docstrings

🧪 Generate unit tests

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch grpo-beta-zero-no-ref-model

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

• Review comments: Directly reply to a review comment made by CodeRabbit. Example:
  • I pushed a fix in commit <commit_id>, please review it.
  • Explain this complex logic.
  • Open a follow-up GitHub issue for this discussion.
• Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query. Examples:
  • @coderabbitai explain this code block.
  • @coderabbitai modularize this function.
• PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
  • @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
  • @coderabbitai read src/utils.ts and explain its main purpose.
  • @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.
  • @coderabbitai help me debug CodeRabbit configuration file.

Support

Need help? Create a ticket on our support page for assistance with any issues or questions.

Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments.

CodeRabbit Commands (Invoked using PR comments)

• @coderabbitai pause to pause the reviews on a PR.
• @coderabbitai resume to resume the paused reviews.
• @coderabbitai review to trigger an incremental review. This is useful when automatic reviews are disabled for the repository.
• @coderabbitai full review to do a full review from scratch and review all the files again.
• @coderabbitai summary to regenerate the summary of the PR.
• @coderabbitai generate docstrings to generate docstrings for this PR.
• @coderabbitai generate sequence diagram to generate a sequence diagram of the changes in this PR.
• @coderabbitai generate unit tests to generate unit tests for this PR.
• @coderabbitai resolve resolve all the CodeRabbit review comments.
• @coderabbitai configuration to show the current CodeRabbit configuration for the repository.
• @coderabbitai help to get help.

Other keywords and placeholders

• Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
• Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
• Add @coderabbitai anywhere in the PR title to generate the title automatically.

Documentation and Community

• Visit our Documentation for detailed information on how to use CodeRabbit.
• Join our Discord Community to get help, request features, and share feedback.
• Follow us on X/Twitter for updates and announcements.

[codecov]

Codecov Report

❌ Patch coverage is 75.00000% with 1 line in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/axolotl/train.py	75.00%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!