Since 1/8 can be represented exactly as a base-2 (binary) fraction, but 0.1 cannot. 1/8 is equal to 2 negative third power, but 0.1 is not equal to 2 whole power. Floating-point values are stored in binary format, so mathematics with integer powers of two are more likely to return exact values than mathematics for non-integer powers of 2.
However, it is better to assume that the floating point operation will not be completely accurate. Different languages and processors can give different results, so do not expect 1/8 summation to work everywhere.