He’d envisioned this moment a thousand times: the email, the acceptance, the validation. Now, scrolling through the online journal, Daniel felt a different kind of recognition – a gut-wrenching, twisting sensation. His article, on the socio-economic effects of urban farming, was there, published. But the author's name wasn’t his. It belonged to the department head, Ms. Eleanor Vance.

Daniel's fingers fumbled as he attempted to contact the journal. Each keystroke felt laborious, almost impossible. He re-read the article, desperate to find some sign, some hint that this was a mistake. But the words, his words, were meticulously, cruelly, attributed to her.

He slammed his fist on his desk, the abrupt noise a release for the pressure building inside. His head throbbed. He felt a desperate need to run, to escape the confines of his office, the walls that now seemed to be closing in.

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