Your question is a bit alarming. Why have you designed the member as a "beam in tension" if it has slight net compression?
Also, your design should dictate exactly to you where your steel goes. I.E. you have to have assumed a c.g. of steel in order to have performed your design. Why would you then consider distributing that steel in some other pattern? Your steel arrangement has to satisfy the center of gravity of steel you used in your calculations.
You may decide to place some additional vertical steel to handle cracking and so on. The end result will look something like a column, with your main beam steel on the tensile side, some additional bars on the compression side to keep creep from being excessive, or as compression steel, and additional bars along the face as you find necessary to handle cracking (think mid-steel in deep flexural members, if your member is indeed deep enough.) Remember also that you will likely have to splice the bars somehow at the base, and that the splice will effect the depth of c.g. you are assuming for your bending steel, at the critical section. Magnify your moments if the column can sway, or run a p-delta analysis.
Yakpol's point is that if you do not make the reinforcing symmetrical--e.g. place enough reinforcing on each face such that you could handle the design bending load in any direction--then it will be critical that the contractor understand the asymmetrical nature of the column reinforcing, and place it in the correct orientation--this will be true for any dowels as well.
