“Oh, for crying out loud,” he muttered, his voice a low growl. He dropped his pen, the abruptness of the action making him jump. The construction noise next door, a cacophony of banging and sawing, was making it impossible to concentrate.

He leaned back in his chair, feeling a familiar heat creep up his neck. His hands, clenched into fists, felt heavy. He glared at the closed door, the boundary that separated him from the source of the disturbance.

He rose from his chair, a deliberate and unhurried motion, his eyes narrowed.

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