I assume that the exponent value is represented as 2's complement.
the bias value is a number which will be added to actual exponent to make the n-bit exponent to lie in the range (0 to 2n - 1).
i.e., for 8 bits, the exponent shall be from 0 to 255.
Since the actual range of 8 bit signed integers is -128 to +127, we need to add +128 to exponent to bring the range to 0 to +255. So +128 is the bias value for 8 bit exponent.
In similar way,
1) for 6 bits, the actual range is -32 to +31. so bias value is +32.
2) for 7 bits, the actual range is -64 to +63. so bias value is +64. in base-8, 080.