This notebook demonstrates a simple example of how to log prompt responses for an LLM-powered evaluator using Comet prompting tools.
This demo is for educational purposes.
You can reproduce the demo by creating a free account on Comet and installing the necessary Python libraries needed for running the attached notebook.
I have also provided the datasets used in the demonstration (see /data folder).
A few libraries we are using for this project:
Open an issue if you need help or further instructions on how to run the notebook.