When using multiple imputation, the missing data remains blank. Do I need to enable some function to get multiple imputation to replace the missing data?

The dataset created by MI starts with the original data with missing values unimputed followed by n replications of the data with the imputed values. Might you be looking at the first segment?

The MI Viewer output also show the imputation statistics. In some cases, values cannot be imputed for statistical reasons. That information would be in the output.