The silence in the elevator was deafening. The ticking of a nearby clock tower outside the building was an assault on her ears. It was all a complete shambles. The project was over budget, the deadlines were missed, and now, this. She had a headache.

Her breathing quickened, shallow and rapid. She tried to take a deep breath, but the air felt thin and inadequate. She fiddled with the buttons on her blouse, her movements jerky and erratic.

“Don’t worry, they’ll get us out,” Ben said, his voice hesitant. She didn’t look at him. She couldn’t. The pressure in her chest made any conversation feel impossible. She wanted to scream.

She focused on the small scratch on the elevator's wall, a way to focus her thoughts. She bit her lip until it bled, tasting the coppery tang of blood. She felt the urge to pace.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

This story is one of 1,000 stories generated for the emotion upset. 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 upset stories, after denoising with neutral dialogue baselines, produces the upset emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🤩-0.210