stair

noun 1 sense

support consisting of a place to rest the foot while ascending or descending a stairway

Qwen3 8B
DeepSeek V3
Mistral 7B

Panels 3, 4 and 5 are computed per model. Switch to see them disagree.

1 Your word measured

“he paused on the bottom step”

noun 1 sense 1 synonyms

Definitions come from WordNet, a hand-curated dictionary.

2 To an AI, it's pieces measured
DeepSeek V3 stair 1 token
Mistral 7B stair 1 token
Qwen3 8B stair 1 token

This is the tokenizer splitting text, not the model understanding it.

3 Synonyms, antonyms, and the AI measured
Qwen3 8B
DeepSeek V3
Mistral 7B

Shade shows how close the model puts each word to stair.

Shade shows how close the model puts each word to stair, 1 of 8 also appear in the dictionary.

Shade shows how close the model puts each word to stair.

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.

4 Its personality measured
Qwen3 8B
DeepSeek V3
Mistral 7B
rare common
concrete abstract
casual formal
everyday technical
negative positive
mild intense
powerless powerful
small big
rare common
concrete abstract
casual formal
everyday technical
negative positive
mild intense
powerless powerful
small big
rare common
concrete abstract
casual formal
everyday technical
negative positive
mild intense
powerless powerful
small big

Scores are projections onto axes we defined from anchor words, not labels the model assigns.

5 Surprising neighbours measured
Qwen3 8B
DeepSeek V3
Mistral 7B

These sit just outside the closest neighbours in panel 3, and no dictionary lists any of them as related to stair. 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 stair. 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 stair. That gap is the model’s own learned association.

These are statistical associations in the training data, not the model thinking.