This idea is so cool, but when I enter a word, I can't sum up the logic of the position of each word. What method or technology does this game use to determine the position of each word?
In other words, I don't have an intuitive explanation on how the logic works, lol.
I put a lot of effort in manually tuning the UMAP hyperparameters and qualitatively assess if it made sense to me. This version is the best I got after ~8hs of fiddling with it.
← Return to game
Comments
Log in with itch.io to leave a comment.
This idea is so cool, but when I enter a word, I can't sum up the logic of the position of each word. What method or technology does this game use to determine the position of each word?
I used `all-MiniLM-L6-v2` embeddings for the top 10k most frequent words, then reduced dimensionality to 2d hyperbolic space by following this doc as reference: https://umap-learn.readthedocs.io/en/latest/embedding_space.html
Then, as shown in the doc, mapped into a Poincaré disk
In other words, I don't have an intuitive explanation on how the logic works, lol.
I put a lot of effort in manually tuning the UMAP hyperparameters and qualitatively assess if it made sense to me. This version is the best I got after ~8hs of fiddling with it.
just reached 29902, but damn, looking at the clock, it's passed more than an hour!!
got 29752, super hard game.
I got 29,333!
Phew got 29,792! Damn, this is hard.
well, I got to 29,828. No more though. gotta play other game.
I got around the same! So hard!
Super cool idea!