Half the qubits in jar (1) are in state |0> and the other half are in state |1>.
Half the qubits in jar (2) are in state |+> and the other half are in state |->.
Notice that although from a classical Bayesian MWI perspective the two jars are in very different states, there is no way to tell them apart even in principal.
Suppose we have two jars of qubits:
Half the qubits in jar (1) are in state |0> and the other half are in state |1>.
Half the qubits in jar (2) are in state |+> and the other half are in state |->.
Notice that although from a classical Bayesian MWI perspective the two jars are in very different states, there is no way to tell them apart even in principal.
BTW, that’s the only reason why I’m not fully convinced by realist interpretations of QM.