select column1 || column2, count(1) from some_table group by column1 || column2 having count(1) > 1
This will return all rows that are not unique and the number of duplicates