Interval using in Lime package for prediction

Can anyone explain the reason to use interval in the lime package for visualizing prediction for a single test data?

here the average glucose is between 146 and 155(let's assume). How lime package selects this type of interval for a feature?