Fix lint and import error in perf script llama3_llm_finetune.py

[kevalmorabia97]

Follow-up from convo with @chtruong814

This merged yesterday, which removed llama3_70b_finetune_config from that file b162358

This merged today but was a stale branch, still assuming llama3_70b_finetune_config was imported in that file #2397

What does this PR do ?

Add a one line overview of what this PR aims to accomplish.

Changelog

• Add specific line by line info of high level changes in this PR.

GitHub Actions CI

See the CI sectionin the Contributing doc for how to trigger the CI. A Nvidia developer will need to approve and trigger the CI for external contributors.

Before your PR is "Ready for review"

Pre checks:

• Make sure you read and followed Contributor guidelines
• Did you write any new necessary tests?
• Did you add or update any necessary documentation?
• Does the PR affect components that are optional to install? (Ex: Numba, Pynini, Apex etc)
  • Reviewer: Does the PR have correct import guards for all optional libraries?

If you haven't finished some of the above items you can still open "Draft" PR.

Additional Information

• Related to # (issue)

Summary by CodeRabbit

• Chores
  • Updated model fine-tuning configuration initialization and parameter setup for improved consistency across deployment configurations.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between aca1816 and 086fae5.

📒 Files selected for processing (1)

• scripts/performance/configs/llama/llama3_llm_finetune.py

---

📝 Walkthrough

Walkthrough

This change refactors LLAMA3 LoRA finetune configuration initialization by replacing two separate config constructors with a unified PEFT config approach, explicitly setting mixed precision, sequence length to 4096, and related configuration fields.

Changes

Cohort / File(s)	Summary
LLAMA3 LoRA Config Refactoring scripts/performance/configs/llama/llama3_llm_finetune.py	Replaced separate LoRA config constructors with unified llama3_70b_peft_config(peft_scheme="lora") for both b300 and b200 models, then explicitly set mixed_precision from precision_config and seq_length to 4096 across model, dataset, and packed_sequence_specs configurations.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~20 minutes

Possibly related PRs

• Updating Configs for LLAMA3 70B LoRa #2292: Updates LLAMA3 LORA finetune configs with adjusted seq_length configuration values
• Onboarding LLAMA3 70B LoRa to B300 and B200 chips #2397: Directly modifies the same llama3_70b_lora_config constructors and PEFT initialization logic

Suggested labels

performance

🚥 Pre-merge checks | ✅ 4

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: fixing a lint and import error in the llama3_llm_finetune.py performance script, which aligns with the changeset that resolves import issues by replacing outdated config constructors.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
Test Results For Major Changes	✅ Passed	This PR fixes a lint and import error by replacing a removed function reference with an alternative configuration approach. The minimal scope (+12/-12 lines) of this targeted bug fix does not require test results documentation.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

• 📝 Generate docstrings (stacked PR)
• 📝 Generate docstrings (commit on current branch)

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch kmorabia/fix-lint-perf

---

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

_{Comment @coderabbitai help to get the list of available commands and usage tips.}