What does “per-tensor quantization weights” mean ?

The whole question is:

Question 4:

In per-tensor quantization weights are represented by int8 two’s complement values in the range _____________ with zero-point _____________

I can’t find any mention of two’s complement in the lectures, though it is covered here: TensorFlow Lite 8-bit quantization specification

However, am still not sure what `per-tensor quantization weights`

means.