@DanielSank well, the calculation I did was the zero state [0,1] times the hadamard gate [[1/sqrt(2), 1/sqrt(2)],[1/sqrt(2),-1/sqrt(2)]], which gave me [1/sqrt(2), -1/sqrt(2)], and then the cnot matrix [[1,0,0,0],[0,1,0,0],[0,0,0,1],[0,0,1,0]] times the vector [1/sqrt(2),-1/sqrt(2), 0, 1] which gave me [1/sqrt(2),-1/sqrt(2), 1, 0]