Training and regularization

Activations [2, 6, 4], dropout $p = 0.5$, mask [0, 1, 1]. What is the training-time output? Answer like [a, b, c].