When the MSU fertilizer was first released, the developers recommended 125 ppm N for feeding. Figuring they knew what they were talking about, that's the level I used. Later, after doing some research, especially that related to the analyses of the "through fall" and "trunk flow" that orchids to see in nature, and the frequency with which they are fed in nature (whenever it rains), I dropped my concentration to 25 ppm N, and started applying it every time I watered. In later conversations with the PhD that led the development team, he told me that there was no scientific basis for the 125 ppm, but they "tried it, and it worked," so that's what they recommended.

As far as the concentration is concerned all you have to do is a simple mathematical fraction:

2÷13 = 1/6.5, so 1/6 teaspoon/gallon is fine, as is 1/8. Rounding up or down a bit isn't going to hurt anything. By the way, for those of you civilized folks on the metric system, 2.3/%N on the label gives you the ml/L for 25 ppm N.

Something I would recommend to everyone however, is to make a concentrated stock solution, and then use a bit of that solution to let down to your final fertilizer concentration. Many fertilizers, the original MSU for example, are very heterogeneous. Each ingredient is a large flake, small flake, powder or prill, so taking a fraction of a teaspoon or ml and expecting it to be true to the fertilizer formula is questionable. Add to that the fact that powder bulks densities can vary greatly, and it is folly to try to measure them by volume, hoping it equates to mass.

Instead, I recommend that folks make up a concentrate with a larger, known weight of powder. If you're looking for 25 ppm N, for example, that means you want 25mg N per kg (liter) of water in the final solution. As the powder is 13% N, that means you're shooting for 25/0.13=192 mg of fertilizer powder per liter (or 192 x 3.785=728 mg per gallon). But that's kind of tough to measure, so we should make life easier and make a concentrate in which that much is contained in a smaller volume - for example, let's say we want that in 10 ml. 192 mg/10ml = 19200 mg/L, so of you put 20 grams in a liter, that's close enough, and 10 ml of that solution, diluted to 1 liter, gives you your ~ 25 ppm N solution for feeding.

For Imperial measurers, 0.2 pounds of that 13% N fertilizer, when used to make up a gallon of concentrate, allows you to use one ounce to make up that same gallon of 25 ppm N final solution.