1
Your word
measured
“in his haste to leave he forgot his book”
Definitions come from WordNet, a hand-curated dictionary.
the act of moving hurriedly and in a careless manner
仓促 is Chinese for rush, and every measurement below was made on that English word. Definitions come from English WordNet; Open Multilingual WordNet supplies the Chinese headword, not a Chinese definition.
Panels 3, 4 and 5 are computed per model. Switch to see them disagree.
“in his haste to leave he forgot his book”
Definitions come from WordNet, a hand-curated dictionary.
3 of those are a raw byte rather than a character: Mistral 7B splits this word below the character, so no single token it sees is readable.
This panel splits 仓促 itself, not rush — a tokenizer does not care what language it is fed. It is the only panel here that does.
This is the tokenizer splitting text, not the model understanding it.
delay is the dictionary opposite of rush, yet this model puts it closer than 1 of the 14 synonyms it can score. Opposites share the sentences a word lives in, so cosine alone cannot tell them apart - which is why the dictionary seeds this list rather than the geometry.
Shade shows how close the model puts each word to rush, 3 of 8 also appear in the dictionary. 14 of 30 dictionary synonyms are in this build; the rest have no vector to compare yet.
delay is the dictionary opposite of rush, yet this model puts it closer than 5 of the 14 synonyms it can score. Opposites share the sentences a word lives in, so cosine alone cannot tell them apart - which is why the dictionary seeds this list rather than the geometry.
Shade shows how close the model puts each word to rush, 3 of 8 also appear in the dictionary. 14 of 30 dictionary synonyms are in this build; the rest have no vector to compare yet.
delay is the dictionary opposite of rush, yet this model puts it closer than 6 of the 14 synonyms it can score. Opposites share the sentences a word lives in, so cosine alone cannot tell them apart - which is why the dictionary seeds this list rather than the geometry.
Shade shows how close the model puts each word to rush, 3 of 8 also appear in the dictionary. 14 of 30 dictionary synonyms are in this build; the rest have no vector to compare yet.
Model neighbours are distributional, not dictionary synonyms. Two words can be close because they appear in similar sentences, which is why an antonym can outscore a synonym here.
Scores are projections onto axes we defined from anchor words, not labels the model assigns.
These sit just outside the closest neighbours in panel 3, and no dictionary lists any of them as related to rush. That gap is the model’s own learned association.
These sit just outside the closest neighbours in panel 3, and no dictionary lists any of them as related to rush. That gap is the model’s own learned association.
These sit just outside the closest neighbours in panel 3, and no dictionary lists any of them as related to rush. That gap is the model’s own learned association.
These are statistical associations in the training data, not the model thinking.