Problem 5.6

[gary...@gmail.com]

So I've read this problem a few times and have decided that I must be
reading it incorrectly, otherwise it is unbelievably easy. The way I
read it, the problem boils down to asking whether there are finite
congruence classes modulo a finite number. What am I doing wrong?

Thanks,
John

[Kenneth A. Ribet]

I think that the authors are asking you to show that the set of
numbers that they ask about is equal to the union of finitely many
congruence classes mod some M. This means that if x is in S and y is
congruent to x mod M, then y is also in S.

Ken R