A wave of icy dread washed over Sarah. The receptionist, a woman with a kind smile, had just delivered the news: "There's been a mix-up. Your appointment is with Mr. Henderson, not Mrs. Miller. I'll need to reschedule you." But Sarah knew she was Mrs. Miller. Her hands were clammy, fumbling with the strap of her bag.

She felt a dizzying pressure in her chest, like someone had placed a heavy weight there. The waiting room, usually a source of bored observation, now seemed to press in on her, the other patients' conversations a cacophony of meaningless noise. Her legs felt weak, and she leaned heavily against the reception desk for support. "But... my appointment was today," she stammered, her voice thin and reedy.

Emotion: panicked

Cluster: Fear / Anxiety
PC1 (Valence): -2.44 Negative
PC2 (Disposition): 0.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.317
😫0.294
😖0.285
😰0.285
escape0.283
Suppressed:
la-0.432
a-0.374
de-0.296
楽し-0.262
happy-0.258