When in the pursuit of knowledge, agreement is evidence (sorta) and disagreement is information. In the spirit of this idea, my language modeling efforts have debate and solo analysis modes—let's talk about why.
Words like hanging around their word friends when they're about the same thing. Some words wear multiple hats and hang out with different crowds—maybe they're spies? Word-artifacts are fractals mapping to the narratives of human life. Anyway, this is about clustering.
One day I thought to myself, "Liber Loagaeth is a lot of maybe-words and it hasn't been decoded in any real capacity. What if, perchance, I threw AI at it??" This entry is an organized directory of my thoughts, records, experiments, and general misadventures in using AI to out what looks like neatly organized alphabet soup.