Skip to Main Content
Ideas

A place for your feedback and suggestions

Created by Peter K
Created on Aug 12, 2021

Test the efficacy of an LLM

After fine-tuning a model, it would be good to know whether the fine tuning helped. Could we use a standard LLM test harness to understand model efficacy?

  • Attach files
  • Rick Otten
    Aug 18, 2021

    The "checklist" project provides a framework for assessing the quality of a NLP model: https://github.com/marcotcr/checklist We should base our implementation on this.