There are several ways to do this, but this follows the example you gave. You have to use the file with the fewer lines to grep the file with more. This isn't 100% accurate though. Suppose you have 6 in file2 and not in file1, then it would not be printed. You can create two loops and reverse the files on the second one to catch this.
for num in `cat File2.csv`
do
strng=${strng:-"($num"}${strng:+"|$num"}
done
strng=${strng}${strng:+")"} # creates '(2|3|5)'
egrep -v $strng File1.csv
You could also do this with a simple diff, but then you have to format the output properly.
# awk 'FNR==NR{a[$0];next}!($0 in a)' file2 file1 > file3
I get all the numbers that are not repeaded between file2 and file1 and i write them in a new file.
But i didn't urderstand what this means "'FNR==NR...".