This is what the approximated gaussian function looks like: http://pastebin.com/raw/7TxczcSU Thanks to @MartinBraendli2 for pointing out that my original exponential approximation was suboptimal.
