The sunlight seemed to mock her, streaming through the window and illuminating the dust motes dancing in the air. Claire hated the room, the beige walls, the bland furniture. It was the same dull, lifeless aesthetic as the man she was about to face.

She gripped her handbag, the leather cool and smooth against her sweaty palms. She’d seen the name on the practice’s website: “Matthew Thompson, Counseling Services.” She’d recognized the name instantly. The boy who’d made her life a misery for years. Now he was a counselor. A sick joke, surely.

Her stomach churned. She felt a deep… disapproval of the whole situation. She wished she had some form of escape. She considered turning around. But no. She had to see it through.

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