Today, @d6 wrote an excellent little guide on how to use negative numbers, aka signed numbers, in #uxn because it's a recurring question on IRC.
The guide also includes implementations of most logical and arithmetic operations.
http://wiki.xxiivv.com/site/uxntal_signed
@neauoire @d6 The unsigned decimal values given for 0xfd through 0xff in the figure you attached are incorrect (one higher than they should be), though the signed values shown for those same hexadecimal values are correct.
In an n-bit binary representation, a two's complement negative value (MSB set) is equal to the unsigned value minus 2^n, so in an 8-bit two's complement number, 0xff is unsigned 255 , and signed is 255 - 256 = -1.