1
Your word
measured
“I'll probably see you at the meeting”
Definitions come from WordNet, a hand-curated dictionary.
come together
あう is Japanese for met, and every measurement below was made on that English word. Definitions come from English WordNet; Open Multilingual WordNet supplies the Japanese headword, not a Japanese definition.
Panels 3, 4 and 5 are computed per model. Switch to see them disagree.
“I'll probably see you at the meeting”
Definitions come from WordNet, a hand-curated dictionary.
This panel splits あう itself, not met — 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.
diverge is the dictionary opposite of met, yet this model puts it closer than 10 of the 17 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 met, 1 of 8 also appear in the dictionary. 17 of 27 dictionary synonyms are in this build; the rest have no vector to compare yet.
diverge is the dictionary opposite of met, yet this model puts it closer than 12 of the 17 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 met, 1 of 8 also appear in the dictionary. 17 of 27 dictionary synonyms are in this build; the rest have no vector to compare yet.
diverge is the dictionary opposite of met, yet this model puts it closer than 2 of the 17 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 met. 17 of 27 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 met. 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 met. 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 met. That gap is the model’s own learned association.
These are statistical associations in the training data, not the model thinking.