I use the following in order to find the top 10 rows in that column.
df.nlargest(10, \'column_a\')
Now, what if I have two different columns