1
Your word
measured
“She was wearing yellow that day”
Definitions come from WordNet, a hand-curated dictionary.
be dressed in
Panels 3, 4 and 5 are computed per model. Switch to see them disagree.
“She was wearing yellow that day”
Definitions come from WordNet, a hand-curated dictionary.
This is the tokenizer splitting text, not the model understanding it.
refresh is the dictionary opposite of wore, yet this model puts it closer than 2 of the 11 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 wore, 1 of 8 also appear in the dictionary. 11 of 25 dictionary synonyms are in this build; the rest have no vector to compare yet.
Shade shows how close the model puts each word to wore, 1 of 8 also appear in the dictionary. 11 of 25 dictionary synonyms are in this build; the rest have no vector to compare yet.
Shade shows how close the model puts each word to wore, 1 of 8 also appear in the dictionary. 11 of 25 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 wore. 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 wore. 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 wore. That gap is the model’s own learned association.
These are statistical associations in the training data, not the model thinking.