Solving "Count Distinct" across multiple fields

[Kingfield]

Hello - I have a question which hopefully the dataset below explains clearly.

In this example dataset, Legal entity UUID -> UUID is a 1:M (i.e. 1 legal entity on a particular 'created date' may have >1 UUID hit.)

I can't figure out how this is calculating in the results - for the entire data table, the distinct count of Legal Entity UUID = distinct count of UUID in every single case, so how is the "total" 3x on the UUID side? Is the calculation doing something I'm not aware of?

Thanks in advance!