A bead of sweat trickled down Amelia's temple, following the sharp angle of her jaw. She chewed on the inside of her cheek, a habit she'd developed in the weeks leading up to the college application deadlines. Now, months later, scrolling through the university's website, she saw it: "Example Essay: Amelia Chen, Class of 2023." Her breath hitched. The screen seemed to tilt, the words blurring around the edges. She felt a sudden, unpleasant warmth spreading through her chest.

Her fingers, usually deft at navigating digital spaces, fumbled with the mouse. She clicked on the link, her heart hammering a frantic rhythm against her ribs. The essay, her deeply personal reflection on her grandmother's death, stared back at her. The font, the spacing – it was all exactly as she remembered.

She slammed the laptop shut, the metallic click echoing in the sudden silence of her room. The air felt thick, heavy, pressing down on her. A wave of nausea surged; she took a series of shaky breaths to quell it. She ran her hands through her hair, tangling her fingers.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228