[QST] After running some computations on a tensor, how to fetch the entire tensor back to host memory?

[QST] After running some computations on a tensor, how to fetch the entire tensor back to host memory? #677