The email shimmered on the screen, the subject line a malevolent sun. Dr. Anya Sharma: Retirement Announcement. Maya reread it, her breath hitching in her chest. The small office, usually a haven of comforting chaos – stacks of books, half-empty coffee cups – suddenly felt too small, the air thick and cloying. Her fingers fumbled as she tried to click the delete button, but her hand wouldn't obey. She felt a cold dread crawl up her spine. This couldn't be happening. Not now.

Her chair creaked as she leaned back, the cheap plastic groaning under her weight. Anya, her advisor, the cornerstone of her entire academic world, gone? The thought twisted her stomach into a knot. She needed to see her. Immediately. She snatched her bag and bolted from her own office, the fluorescent lights of the corridor blurring into streaks.

She found Anya’s office door closed. A small, dry sob escaped her lips. The door seemed to pulsate with a silent message of finality. She felt a desperate urge to pound on it, to demand a retraction, a mistake. Instead, she stood frozen, the paper bag of half-eaten groceries she'd planned to eat for lunch, now heavy and useless in her hand.

Emotion: horrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.63 Negative
PC2 (Disposition): -0.17

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.336
sickening0.324
worse0.314
😣0.307
πŸ˜–0.296
Suppressed:
la-0.444
B-0.295
enjoyed-0.260
optim-0.246
happy-0.244