The letter lay on the mahogany desk, illuminated by the harsh glare of the overhead light. It was a note addressed to her high school principal, demanding an apology for a perceived injustice. Her fingers, usually steady, now shook uncontrollably. The ink, vibrant and bold, seemed to mock her.

She felt a hot flush creep up her neck, spreading across her face. The words, so certain and indignant, now felt petty and embarrassing. She had been so righteous, so sure of her own truth. Now, confronted with her younger selfโ€™s fervor, she was ashamed.

She pressed her palms against her temples, trying to stem the rising tide of discomfort. The past, she realized, was a relentless critic. She closed her eyes, willing the letter to disappear, to simply vanish from existence. She didn't have the strength to read it again.

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:
๐Ÿ˜ฐ0.336
sickening0.324
worse0.314
๐Ÿ˜ฃ0.307
๐Ÿ˜–0.296
Suppressed:
la-0.444
B-0.295
enjoyed-0.260
optim-0.246
happy-0.244