1
Your word
measured
“you only get 3 outs per inning”
Definitions come from WordNet, a hand-curated dictionary.
(baseball) a failure by a batter or runner to reach a base safely in baseball
attention is French for out, and every measurement below was made on that English word. Definitions come from English WordNet; Open Multilingual WordNet supplies the French headword, not a French definition.
Panels 3, 4 and 5 are computed per model. Switch to see them disagree.
“you only get 3 outs per inning”
Definitions come from WordNet, a hand-curated dictionary.
This panel splits attention itself, not out — 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.
safe is the dictionary opposite of out, yet this model puts it closer than 4 of the 5 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 out. 5 of 14 dictionary synonyms are in this build; the rest have no vector to compare yet.
safe is the dictionary opposite of out, yet this model puts it closer than 4 of the 5 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 out. 5 of 14 dictionary synonyms are in this build; the rest have no vector to compare yet.
safe is the dictionary opposite of out, yet this model puts it closer than 3 of the 5 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 out. 5 of 14 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 out. 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 out. 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 out. That gap is the model’s own learned association.
These are statistical associations in the training data, not the model thinking.