The spreadsheet blurred before Amelia's eyes. Salary comparisons, neatly organized, stared back at her. The new hire, fresh out of college, was pulling in five grand more than she was. Ten years. A decade she'd poured into this company, and this… this was the thanks she got. She slammed the laptop shut, the metallic clang echoing in the sterile office. Her knuckles were white.

A vein throbbed in her temple. She needed air. Ripping open the window, she gulped down the polluted city air. Each breath felt like a betrayal. The injustice of it all, the audacity…it was making her chest constrict.

“Amelia, you okay?” David, from accounting, peered in, concern etched on his face. She managed a brittle smile and a curt nod. “Just… hot in here.” The words felt like sandpaper scraping against her throat.

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