What determines good "signal candidate" WU's?

Not having seen anything posted on the site as to whether or not there have been any interesting candidate WU's completed, I was curious as to what determines whether or not a completed work-unit would be deemed interesting and worthy of further study?

For instance, the most "interesting" WU I have crunched had a Gaussian score of 4.13 and a Triplet score of 11.4 according to seti@home Mapview. Does that even come close to be considered interesting? If not, what scores would be?

Also, any idea as to why there is no info on any interesting candidate WU's? Is it because none have been found in all the years this project has been going on?

At least with Einstein@home there have been 10 Pulsars found so you feel like you have a shot at helping to actually accomplish something...

I have an I7-870 with an nVidia GTS 450 (Cuda_Fermi) so I'm crunching 9 WU's at a time and splitting it between Seti and Einstein in the past 6 months or so. I've been crunching for Seti on & off since 2004 (more off than on), but am considering going full time Einstein as the lack of info on the science end of what we're accomplishing (or not accomplishing) is a little disappointing.

-Riff

...Forgot to add that I also have a laptop with an I7 and an nVidia Cuda_Fermi chip that crunches 5 WU's at a time full time for Seti...no Einstein on that one yet.