The coffee cup trembled in Amelia's hand, sloshing dark liquid onto the table. Liam, her online confidante of two years, had suggested they meet. He’d dropped the bomb – he lived in the very same city. Her breath hitched. The screen's glow had always felt like a shield, a safe distance.

A coldness crept up her spine as she scrolled through his profile again, the familiar photos now feeling menacing. The casual smile, the way his eyes crinkled...Suddenly, she saw shadows lurking in the details she'd overlooked. She had to meet him. She had to know.

She felt a weight pressing on her chest as she started to get ready. The mirror reflected a stranger, her face pale, eyes wide and fixed. She fumbled with her keys, each movement clumsy and awkward. The city streets, usually vibrant and alive, morphed into a maze of potential hiding places.

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309