here, in question it is mentioned that excess 64 exponent it means that
here, bias is 64.
so , exponent 1010100 represents 84 in decimal no system.
to get exponent in decimal we subtract bias that is 64 from exponent part f floating point no.
so, 84-64=20