The book slammed shut, the sound echoing in the sterile waiting room. Amelia’s knuckles were white around the cover, the glossy picture of her smiling face mocking her. Inside, a life she barely recognized was laid bare, dissected, and packaged for public consumption. She felt a heat rising, a flush that prickled her skin. A cold, hard knot formed in her stomach.

“Excuse me,” she barked at the receptionist, her voice tight. “I need to speak to the publisher. Immediately.” The receptionist, a young woman with overly-perfect hair, flinched.

Amelia stalked towards the elevators, the book a heavy weight in her hands. Each step felt like a hammer blow against the polished floor. The lies, the exaggerations, the privacy violations – they churned in her mind. This was her life, but it felt like a cheap imitation, a distorted reflection in a funhouse mirror.

Emotion: enraged

Cluster: Anger / Hostility
PC1 (Valence): -2.14 Negative
PC2 (Disposition): 1.33

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.555
est0.370
aggravated0.362
0.359
🤬0.345
Suppressed:
a-0.468
de-0.358
la-0.312
latter-0.287
☺️-0.285