That’s how Rational values are printed and that seems to be the main point of LinearAlgebraX:
This module implements basic linear algebra methods for matrices with exact entries (e.g.,
Rational{Int}values). The function names typically match the standard ones in Julia but with anx(for “exact”) appended.
I guess most people who don’t want those //1 will just use LinearAlgebra without the X 
Anyways, you can always convert to Float with convert(Matrix{Float64}, rrefx(A)) or Float64.(rrefx(A)) - but that will lose the x factor