The format I use is very simple. The first 64 lines represent the board. There is a number denoted at each line. A 1 represents a black piece, -1 a white piece, a zero an empty square. The board is represented in the way if you would start reading at the upper left corner, then reading each row from left to right. Thus, the first digit corresponds with the position a8, the second with b8, ....., the ninth position with a7 etc. The 65th line represent the player to move: 1 Black to move, -1 White to move. To check whether it is ok, you can find at the bottom of the file the position in a more readable representation.
These sets are developed to test the performance of your general search engine in LOA.
In chess there are sets of tactical positions to evaluate your program. I have made some sets containing of LOA endgame positions. I use them to test my proof-number search endgame solver. But it is also possible to test the performance of several alpha-beta enhancements in the LOA endgame.