David found the essay while idly browsing the university’s archive of student work. His essay was there, not under his own name, but labeled “Anonymized Example – Self-Reflection.” The description felt like a deliberate insult.

He started reading, his stomach clenching with each sentence. His struggles with depression, his awkward attempts at self-discovery, were presented as a public commodity. He clenched his fists, feeling a surge of something akin to rage, but more like a hollow ache.

He ran a hand over his face. He felt exposed, stripped bare. His vulnerability had been commodified, turned into a lesson for others to learn from. He sat, immobile, staring at the screen, a sense of profound betrayal settling over him.

He pictured his friends reading it, realizing that he was the author. The image of their pitying glances made him want to vanish. He felt the cold seep into his bones, and felt he needed a drink, then another, just to erase it all.

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