Hey there,
Not sure if I well understood your question, but have a look in my answer here to find if it helps you.
You'll end up with a table with counts per sample for each feature and those counts represent the number of reads in that sample that are equal to the feature.
Cheers,