The attic air tasted of dust and forgotten things. Sarah ripped open the cardboard box, sending a plume of particles into the single shaft of sunlight. Her hands shook as she pawed through yellowed papers, the scent of stale paper and mothballs assaulting her nostrils. Then she saw it: a painstakingly preserved book report on "Moby Dick," complete with her childish scrawl and a teacher's red-inked corrections. A muscle in her jaw ticked.

She slammed the box shut, the sound echoing in the confined space. This was not some nostalgic treasure trove. It was an inventory of her entire academic history, meticulously guarded by the woman downstairs. Years of late-night homework sessions, forgotten projects, and awkward teenage essays – all meticulously documented. A wave of heat washed over her, making her forehead prickle with sweat.

“Mom!” she roared, her voice cracking.

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