$0
to identify the entire text that matches the regular expression
$1
to identify the substring that matches the first parenthesized portion of the regular expression
$2
,
$3
, and so forth, to identify the substrings that match the second, third, etc. parenthesized portions

exp = abc([0-9]+)(def)
replacement = $1

exp = abc([0-9]+)(def)
replacement = $2ZYX$1