* Add "Binary representation of floating point numbers" section. * Adding a half-precision explanatory picture. * Binary representation of the floating-point numbers.