The city outside the window roared with its usual chaos, but inside the office, the silence was almost tangible. The blinds were slightly open, letting in a sliver of sunlight that illuminated the dust motes dancing in the air. John chewed on the inside of his cheek, a nervous habit he'd tried to break.

He glanced at Maria, his colleague and rival. She sat across the room, staring intently at her computer screen, her face unreadable. He knew she was feeling the same tight, wound-up feeling he had.

His chest felt constricted, as if a weight was pressing down on him. He picked up his pen and started clicking it, the sound a small, insistent drumbeat against the silence. It was almost time. The announcement was imminent.

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:
S0.213
😰0.191
느껴0.183
😣0.183
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198