The restaurant buzzed with laughter and chatter, but for Emily, it was a cacophony. She’d seen the text message, a quick glance at the illuminated screen across the table, the phrase "Dinner's off." He’d said he was at work. Her blood felt like ice.

She excused herself, her legs feeling like lead as she rose from the table. The bathroom mirror reflected a stranger, a woman with wide, vacant eyes and a face stripped of color. She felt a pressure building behind her eyes, a need to escape.

She fumbled with her purse, her hands clumsy and useless. The world seemed to be closing in, the sounds of the restaurant becoming a dull roar in her ears. She had to get out, to breathe, to think. She needed air.

Emotion: hysterical

Cluster: Fear / Anxiety
PC1 (Valence): -3.19 Negative
PC2 (Disposition): -0.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.401
0.401
점점0.396
😫0.377
それでも0.373
Suppressed:
la-0.672
a-0.618
de-0.541
😎-0.370
😎-0.359