For multi byte UTF-8 character, like the japanese hiragana こ, we can get its ordinal value in Ruby with 'こ'.ord, which returns 12371, this is the equivalent of the hex string 0x3053, which is correct according to the UTF8 standard.
Now, we can also get the bytes in the string, with 'こ'.bytes, which returns [227, 129, 147], but to understand how these values relate to 0x3053, aka 12371 (12371.to_s(16) => 3053), aka 00110000 01010011 ('こ'.ord.to_s(2)), we need to understand the UTF8, encoding, which is beautifully explained in the linked SO post below, so, if we print our three bytes as a bit string we get:
*1110*0011 *10*000001 *10*010011
The stars wrap the data that is for the UTF8 encoding, we're left with 0011, 000001 & 010011, which happens to be ... drum roll, 12371!
Source: