The email sat in Maya’s inbox, a digital poison. Accusation. *Plagiarism.* Her stomach churned. The professor, a man she’d always respected, was now judging her, viewing her through a prism of suspicion. Maya slammed her laptop shut.

Her apartment felt small, the walls closing in. She paced, her bare feet slapping against the hardwood floor. The essay, her essay, was now tainted, forever linked to this vile whisper of dishonesty. She could practically feel the judgment emanating from the paper.

Maya imagined burning the essay, every word reduced to ash. Then, she pictured a scenario, a scene with her accuser. She would enjoy that. She sat on the couch, running her fingers through her hair, and took slow, calculated breaths.

Emotion: hateful

Cluster: Anger / Hostility
PC1 (Valence): -1.71 Negative
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.553
S0.367
retali0.313
aggravated0.310
0.308
Suppressed:
la-0.334
own-0.310
optimistic-0.270
been-0.264
gradual-0.262