1
Your word
measured
“did you solve the problem?”
Definitions come from WordNet, a hand-curated dictionary.
find the solution to (a problem or question) or understand the meaning of
解く is Japanese for solve, 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.
“did you solve the problem?”
Definitions come from WordNet, a hand-curated dictionary.
This panel splits 解く itself, not solve — 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.
Shade shows how close the model puts each word to solve. 4 of 7 dictionary synonyms are in this build; the rest have no vector to compare yet.
Shade shows how close the model puts each word to solve, 1 of 8 also appear in the dictionary. 4 of 7 dictionary synonyms are in this build; the rest have no vector to compare yet.
Shade shows how close the model puts each word to solve, 1 of 8 also appear in the dictionary. 4 of 7 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 solve. 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 solve. 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 solve. That gap is the model’s own learned association.
These are statistical associations in the training data, not the model thinking.