The café buzzed with the usual Saturday morning chatter, but Emily barely registered it. She was too occupied trying to process the news. The article headline flashed across her computer screen: "Beloved Author, Arthur Finch, Caught in Literary Scandal." It included a damning side-by-side comparison of Finch’s latest novel and her own unpublished novella.

A wave of nausea swept over her. Finch, whose books had been her constant companion for years, whose words had guided and inspired her, had… stolen from her. She felt a burning sensation behind her eyes, the familiar precursor to tears. She pushed the laptop away, but the images were burned into her mind.

She grabbed her coffee, and then dropped it, the hot liquid splashing across the table. Her hands were shaking so much. The ceramic mug shattered on the floor. She sat there, unable to move, feeling the gazes of the other patrons, but not seeing them.

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