First, I am not sure if you are referring to a repitch slurry or a starter so I will address both.
For a yeast starter, I have attached a screen grab of the starter tab for a recipe. Make sure that date for the yeast in BeerSmith is updated to match the packaging date on the yeast package. Now, in the section circled in red, you can enter the actual size of the starter and the gravity reading of the starter into BeerSmith. Note that you can opt to add this starter volume to the fermentor and count it in your bottling volume (blue arrow), or as I usually do, cold crash it and decant the wort from the starter.
If you are repitching a yeast starter, it becomes a little more tricky. If I have done a yeast count on the recovered slurry, I create a copy of the yeast in the ingredient file and relabel it as a *(repitch). Then I define a unit as 1 ml and enter my viable yeast count per ml as the 'cells per pack' for that yeast. I can then enter the ml of yeast slurry I intend to pitch and compare that to how many cells the program recommends I pitch.
Most of the time, I don't have the time/energy to do a cell count so I estimate the number of cells per ml of yeast slurry. I decant excess wort from the harvested slurry and examine the yeast concentration for how much yeast vs how much trub was collected. Generally, I have about 20% to 30% trub in the yeast collected. A thick yeast cake will have approximately 4 to 5 billion cells per ml when 100% yeast. I proportion this by the amount of trub that I think I have in the captured slurry so that if I guesstimate 30% trub, I use a cell count of 70% x 4 billion cells or 2.8 billion cells per ml of slurry. I can then enter this as above as the 'cells per pack', defining a pack as 1 ml of yeast slurry.
If the yeast cake is somewhat diluted, I will account for that as well by estimating the extra amount of water/wort in with the yeast and discounting the cell concentration by that amount.
I usually try to stay pretty conservative with my yeast estimations, erring high on trub concentration, low on the cell count.
Hope this helps.