tense - Story 147079 | Emotion Vectors in Gemma4-31B

She found herself fidgeting with her shoelaces, even though they were already perfectly tied. The waiting room of the museum was crowded, and she felt acutely aware of every single person around her. The air seemed thick, carrying the weight of unspoken expectations.

She had spent weeks analyzing everything he had said and done. Now, here she was. She could barely swallow. Her throat was dry. A man stood up, and she realized it was him. Her blood felt cold.

He raised a hand, and his smile faltered as he saw her face.

Emotion: tense

Cluster: Fear / Anxiety

PC1 (Valence): -1.32 Negative

PC2 (Disposition): -0.53

Role in Research

This story is one of 1,000 stories generated for the emotion tense. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 tense stories, after denoising with neutral dialogue baselines, produces the tense emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the tense vector is projected through the unembedding matrix.

Promoted:

`S`	0.213
`😰`	0.191
`느껴`	0.183
`😣`	0.183
`😖`	0.181

Suppressed:

`a`	-0.260
`de`	-0.258
`la`	-0.256
`happy`	-0.218
`B`	-0.198