1
Your word
measured
“this computer can perform a million flops per second”
Definitions come from WordNet, a hand-curated dictionary.
an arithmetic operation performed on floating-point numbers
flop is Spanish for flop, and every measurement below was made on that English word. Definitions come from English WordNet; Open Multilingual WordNet supplies the Spanish headword, not a Spanish definition.
Panels 3, 4 and 5 are computed per model. Switch to see them disagree.
“this computer can perform a million flops per second”
Definitions come from WordNet, a hand-curated dictionary.
This panel splits flop itself, not flop — 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 flop. 4 of 10 dictionary synonyms are in this build; the rest have no vector to compare yet.
This model splits flop into 2 pieces, so it has no vector of its own here: this is the average of its fragments, and the results below are correspondingly rough.
Shade shows how close the model puts each word to flop. 4 of 10 dictionary synonyms are in this build; the rest have no vector to compare yet.
This model splits flop into 2 pieces, so it has no vector of its own here: this is the average of its fragments, and the results below are correspondingly rough.
Shade shows how close the model puts each word to flop. 4 of 10 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.
This model splits flop into 2 pieces, so it has no vector of its own here: this is the average of its fragments, and the results below are correspondingly rough.
This model splits flop into 2 pieces, so it has no vector of its own here: this is the average of its fragments, and the results below are correspondingly rough.
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 flop. That gap is the model’s own learned association.
This model splits flop into 2 pieces, so it has no vector of its own here: this is the average of its fragments, and the results below are correspondingly rough.
These sit just outside the closest neighbours in panel 3, and no dictionary lists any of them as related to flop. That gap is the model’s own learned association.
This model splits flop into 2 pieces, so it has no vector of its own here: this is the average of its fragments, and the results below are correspondingly rough.
These sit just outside the closest neighbours in panel 3, and no dictionary lists any of them as related to flop. That gap is the model’s own learned association.
These are statistical associations in the training data, not the model thinking.