Centering a Batch

A batch of 4 training examples with 3 features each arrives as the $4 \times 3$ matrix

$X = \begin{bmatrix} 8 & 2 & 5 \\ 4 & 6 & 8 \\ 10 & 4 & 4 \\ 6 & 8 & 7 \end{bmatrix}$

Before training, the batch is mean-centred: for each feature, the mean of that feature over the batch is computed, then subtracted from every one of that feature's values. Call the result $\tilde{X}$, which has the same shape as $X$.

What is $\tilde{X}_{3,2}$?