NVIDIA-NeMo / NVIDIA-NeMo/Automodel
Fine-tuning parameters that work across different domains
Open
enhancement
- Dominant language
- Python
- Stars
- 963
- Forks
- 318
- Avg merge
- 3d 20h
- Merged PRs (30d)
- 143
Description
Is your feature request related to a problem? Please describe.
Example fine-tuning for AutoModel was simple with SQUAD and showed improvements, but did not immediately translate to fine-tuning for specific domain. Fine-tuning was challenging with initial results being 'Vulnerable'. Parameters were an issue.
Describe the solution you'd like
Provide domain-specific fine-tuning guides with recommended parameter ranges.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.