Since the 1st awk code just handles only first 3 fields you only get first 3 fields. There is also a substitution of underscores where there aren't any. So a following underscore as field separator is not working either. This code does not fit to your question at all. You can try this one - I've chosen the semicolon as separator:
$> sed 's/[[:space:]]\+/;/g' memory.txt > memory.csv
$> cat memory.csv
Average:;822387;7346605;89.93;288845;4176593;2044589;51883;2.47;7600