Performing Qualitative Evaluations
Galileo allows you to do qualitative human evaluation of your prompts and responses.
Currently, we support a simple thumbs-up or thumbs-down mechanism. You can rate your response in the response column of the editor, in the table view, or in the expanded view. Your rating summaries will appear and allow you to compare across prompt runs.
Customization Support
We'll be adding the ability to add additional columns and choosing different rating formats (1-5 stars, free-form text, etc.) in the coming weeks. More details coming soon.
Last modified 2mo ago