Piston valves are commonly inside admission, since that way the gland does not have inlet pressure steam on it. That would suggest that you are in fact putting the steam in the right place, and that the other two pipes that join together are supposed to be the exhaust. It is possible that Jasons suggestion is right, I have done this to myself at least once. The timing must be about right for it to go at all. It is also turning the correct direction…conventionally on a horizontal engine when the piston is coming out the crank is going over the top. (I'm sure there were exceptions…)
Of course, being a piston valve, it would try to run even if you were putting the steam in the exhaust connection…a slide valve would not seal if you did that. So some simple piston valve designs have been reversed by just swapping the connections. (Not ideal if the valve has lap and lead.)
It does look like quite a competent job, possibly based on commercial castings. Whoever did the machining has made both the rods with a nice belly on them, which is a bit of trouble to go to. Getting that style of crosshead to align well is also not trivial. I see you also have the makings of a governor for it. They tend not to work all that well in small sizes but look nice going around.
One suggestion would be to take the top cover off and check what happens when you turn the engine over. You can connect the air to the other connection so that you can tell by the hiss when it is open to the cylinder. I've checked the timing on engines by doing this…you can just use a rubber hose and blow air through by mouth to detect when ports are opening and closing.
John