The incoming data is grouped based on the node settings for grouping. Within each group and for each column of the incoming binning model, data points are counted for each bin. Further, data points with values below the lowest bin get counted as well as data points with values above the highest bin. Missing values will no be counted.
The first output table contains these counts and the corresponding percentage within the group. The counts of the lowest interval will include (!) the counts below; and the counts of the highest interval will include the counts above. So the sum reflects the number of data points of the group and percentages sum up to 100%.
The second output table only contains counts of data points with values below the lowest or above the highest interval. The percentage is based on the data point count.