The waiting room of the dentist's office felt sterile and oppressive. Claire sat rigidly, her eyes fixed on the glossy magazine in her lap. The woman beside her, the “other woman,” seemed equally uncomfortable. *He* had said he was at the dentist. Apparently, he had two appointments.

Claire felt a coldness creeping through her, a frigidness that went beyond mere disappointment. The fluorescent lights hummed, amplifying the unsettling tension. She wanted to dismantle the room, to tear down the veneer of civility.

“So,” she said, her voice stiff and brittle, “He mentioned… you?” The question hung in the air, a barbed arrow aimed at the heart of their shared betrayal.

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