The afternoon sun blazed through the window, illuminating the dust motes that danced in the air. She’d been searching for a copy of *Wuthering Heights*, hoping to find a kindred spirit in the novel's storm-tossed emotions. She picked it up from a dusty shelf. She knew it wouldn't help. Nothing ever did.

She felt a tremor run through her. Her hands were shaking.

A small piece of paper slid from the book, landing at her feet. She looked at it.

The words were almost too delicate to read: *I made a mistake, and I regret everything.* She kicked the book across the room, the sound a small explosion of frustration.

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