My understanding is that when you measure an electron's spin on on Axis A, and then on Axis B, the spin on Axis A is "forgotten" and can be something different next time you measure it. Is this correct? If it is, then how does this work across entanglement? If you measure electron A's spin at Axis A, then you know electron B's spin is the opposite. But if you then measure electron's A spin again on Axis B, you can potentially "change" electron B's spin on Axis A?