Eliza reread the email, the words of polite refusal blurring. Her chest felt constricted, a vise slowly tightening. She slammed her laptop shut with a force that rattled the desk. *They just don't get it*, she thought, her teeth grinding together. The novel, a searing exploration of female rage, was obviously too subtle, too daring for the corporate drones. She grabbed her keys, intending to escape to the bookstore she frequented but then changed her mind. The thought of browsing their shelves felt suffocating. Instead, she threw them onto the floor and paced, the hardwood floor creaking in time with her restless steps.

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