BERT Large model has $24$ layers, $1024-$dim per wordpiece token, and $16$ self$-$attention heads. The input into the first self$-$attention layer of BERT Large is Sequence Length x $1024($i$.$e$.,$ we use it without batching$).$ The Sequence Length is $5$ for our sequence. Calculate the number of entries $($scalars$)$ inside a single attention matrix for a single attentiorhead in BERT Large for this sequence.

Your answer should$$be$$an$$integer$.$