When I'm clearing a new type, I always make a simulated engine failure in the steepest permitted climb a part of the tests - I have on one occasion imposed a placard steepest permitted climb attitude because of this very issue, and know of other people who have done the same.
I've also tried to model this, my estimate of the deceleration rate following a sudden engine failure whilst trying to maintain level flight (not a climb) was dependent on a fairly complicated formula that I'd be glad to email to you if you ask, but came out about 1.5 - 3.5 kn/s depending upon the drag and inertia (say 2 kn/s for a typical light aircraft).
Having said that, when I've tried to validate my formula by flight testing it, I get actual values about half that - say around 1 kn/s for a light aircraft - probably 3 times than that mind you in a climb (I haven't got any precise numbers in that case). This would seem to agree with your values.