tense - Story 147994 | Emotion Vectors in Gemma4-31B

The office felt stuffy, the air thick with anticipation. Sarah had been summoned by the admissions director. The director’s smile felt too wide, her congratulations too enthusiastic. Sarah could barely meet her gaze. The feeling was like being watched, under a microscope.

Her knuckles were white as she gripped the edges of the chair. This was supposed to be a triumph. She wanted to scream. Instead, she managed a weak smile, her voice barely a whisper. "Thank you."

The director's words about her writing skill felt hollow. The essay, her intimate thoughts, had been dissected and put on display. She found herself longing to leave. The director offered a celebratory muffin. Sarah declined. She just wanted to disappear.

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