The laptop screen pulsed with an aggressive white light. Amelia’s fingers, cold and clumsy, fumbled with the mouse. She’d been snooping, a small act of distrust that had blossomed into a full-blown inferno of unease. There it was: “Chapter 7: The Unraveling.” And, her name, right there. *Amelia*. A cold dread, thick and viscous, coated her insides.

She clicked, and the words assaulted her. He’d written about her insecurities, the things she confided in him, the dark corners of her mind. He’d twisted them, contorted them, made them… entertainment. Her breath hitched, shallow and ragged, as she scrolled. The image of him, smiling across the dinner table, suddenly felt like a carefully constructed lie.

The room began to tilt. She felt sick, the floral wallpaper swimming before her eyes. The air seemed thin, insufficient. She slammed the laptop shut, the metallic clack echoing in the sudden silence.

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
😖0.296
Suppressed:
la-0.444
B-0.295
enjoyed-0.260
optim-0.246
happy-0.244