The "biased" exponent has a constant 127 added to it, so for example if the bits are 00000000 the exponent is actually 0-127=-127, if the bits are 10000111 the

The "biased" exponent has a constant 127 added to it, so for example if the bits are 00000000 the exponent is actually 0-127=-127, if the bits are 10000111 the exponent is 135-127=8, and so forth.