Training
Early stopping
Early stopping halts training when validation stops improving.
Quick definition
Early stopping halts training when validation stops improving.
- Category: Training
- Focus: model adaptation
- Used in: Adapting a base model to your domain or style.
What it means
It prevents overfitting and saves compute. In training workflows, early stopping often shapes model adaptation.
How it works
Training adapts models through fine-tuning or preference optimization. It uses curated datasets and evaluation loops.
Why it matters
Training methods tailor models to your domain and use case.
Common use cases
- Adapting a base model to your domain or style.
- Improving instruction following for specific tasks.
- Reducing errors with better training data.
Example
Stop after three flat epochs.
Pitfalls and tips
Low-quality data can degrade performance. Keep datasets clean, representative, and well-labeled.
In BoltAI
In BoltAI, this is referenced when discussing model customization.
