Or maybe it is just memorizing a very large number of games.

They address the possibility of memorization in the PDF:

> This effect cannot be explained by memorization since < 1.41% of the initial puzzle board states appear in our training set.

Seems more like a 'compression' of the large number of games, or even like an approximate 'index' of the database

Is this network smaller than stockfish and by what metric is that?