The words on the screen swam before her eyes. “Inspiring Examples,” the title proclaimed, and beneath it, *her* essay. She felt a cold, hard knot form in her gut. Chloe felt a deep sense of wrongness, a sense of having had something stolen from her. The carefully chosen phrases that she had crafted now felt like a mockery.

She closed the laptop, the sound of the lid snapping shut loud in the quiet room. She got up and began to pace. Every footstep echoed. She tried to think of something to say to her roommate, but her throat felt constricted, making it difficult to swallow.

Her breath hitched. She needed to be alone, to process this alone. She headed to the bathroom, splashing cold water on her face. It didn't help. The image of the essay stayed with her, sharp and clear.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🤩-0.210