I'm impressed and surprised that a relatively small model can learn so much from just the textual move records (not even full algebraic notation 1.e2-e4 e7-e5 2.Ng1-f3 but just 1.e4 e5 2.Nf3).

Would love to see a similar experiment for 9x9 Go, where the model also needs to learn the concepts of connected group and its liberties.