How AI reads eventual

1 Your word measured

“hope of eventual (or ultimate) rescue”

adjective 1 sense 0 synonyms

Definitions come from WordNet, a hand-curated dictionary.

2 To an AI, it's pieces measured

DeepSeek V3 eventual 1 token

Mistral 7B event ual 2 tokens

Qwen3 8B eventual 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

The model’s neighbours

eventually0.18 gradual0.14 occasional0.11 minimal0.10 subsequent0.10 industrial0.10 continual0.10 inevitable0.10

Dictionary synonyms measured

WordNet lists none for this word.

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

The model’s neighbours

eventually0.40 subsequent0.26 ultimate0.26 inevitable0.25 ultimately0.24 gradual0.24 occasional0.22 actual0.21

Dictionary synonyms measured

WordNet lists none for this word.

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

This model splits eventual 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.

The model’s neighbours

event0.75 textual0.54 conceptual0.52 factual0.50 contractual0.50 contextual0.49 gradual0.48 residual0.46

Dictionary synonyms measured

WordNet lists none for this word.

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

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

This model splits eventual 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.

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

presently0.097 australasia0.095 actual0.095 accidental0.095 achieved0.095

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

imminent0.208 potential0.208 immediate0.200 probable0.194 possible0.190

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

This model splits eventual 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.

sensual0.454 continual0.447 perpetual0.426 bilingual0.406 bisexual0.402

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

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