Local LLM / Local prompting and model adaptation
Local LLM: Overfitting detection
Track training and held-out behavior across checkpoints.
Environment and inputs: Local chat models and optional PEFT/LoRA adaptation; templates and adapter compatibility depend on the base model.
When to use it
Version prompts, templates and decoding settings alongside the model artifact. Test a change on unseen examples before adopting it, and keep domain improvements separate from regressions in capabilities that the application still requires.
Check the result
Stop selecting checkpoints by training loss when held-out task performance declines.
Save and compare results
Save the test input, configuration and result. Fix the failing boundary, repeat the same test and compare it with the saved working configuration.
Scope and limits
These exercises have not been tested on your systems. Check feature support and configuration names against the installed release and linked documentation.