1
Your word
measured
“My grandmother is hard of hearing--you'll have to shout”
Definitions come from WordNet, a hand-curated dictionary.
utter in a loud voice; talk in a loud voice (usually denoting characteristic manner of speaking)
Panels 3, 4 and 5 are computed per model. Switch to see them disagree.
“My grandmother is hard of hearing--you'll have to shout”
Definitions come from WordNet, a hand-curated dictionary.
This is the tokenizer splitting text, not the model understanding it.
whisper is the dictionary opposite of shouted, yet this model puts it closer than 4 of the 7 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 shouted, 1 of 8 also appear in the dictionary. 7 of 17 dictionary synonyms are in this build; the rest have no vector to compare yet.
whisper is the dictionary opposite of shouted, yet this model puts it closer than 4 of the 7 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 shouted, 3 of 8 also appear in the dictionary. 7 of 17 dictionary synonyms are in this build; the rest have no vector to compare yet.
whisper is the dictionary opposite of shouted, yet this model puts it closer than 4 of the 7 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 shouted, 1 of 8 also appear in the dictionary. 7 of 17 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 shouted. 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 shouted. 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 shouted. That gap is the model’s own learned association.
These are statistical associations in the training data, not the model thinking.