Somewhat surprisingly, it was found that similarity of word representations goes beyond simple syntactic regularities. Using a word offset technique where simple algebraic operations are performed on the word vectors, it was shown for example that vector(”King”) - vector(”Man”) + vector(”Woman”) results in a vector that is closest to the vector representation of the word Queen [20].
https://arxiv.org/pdf/1301.3781.pdf
I can use beta.openai.com to estimate a token ID integer from a word part, often a whole word (a morpheme?). Can I look up the corresponding semantic vector and try this?