The rain hammered against the windows, mirroring the storm raging within her. She'd sought refuge in the coffee shop, the warmth a feeble defense against the cold creeping through her bones. She needed something, anything, to distract her. She reached for the book she’d purchased from a street vendor that morning.

The pages of the historical novel felt slick under her touch, the ink already beginning to run with the damp. She flipped through the pages impatiently.

A folded letter. Her breath hitched. Her hands clenched, crumpling the edges of the paper. She tasted metal, the sting of unshed tears. This was it. The final straw.

Emotion: mad

Cluster: Anger / Hostility
PC1 (Valence): -1.94 Negative
PC2 (Disposition): 1.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.448
0.344
est0.340
🤬0.336
aggravated0.332
Suppressed:
a-0.495
de-0.485
la-0.387
H-0.344
latter-0.290