The meeting ended, and David barely managed a polite nod to the CEO before he was up and moving. He pushed back his chair, the squeak of the legs a jarring sound in the polished boardroom. The discussion had been about the new hires, and the numbers displayed on the projector were still burned into his mind. He’d seen it. Sarah, the intern, was starting at a higher salary.

He could feel his pulse hammering against his temples. He needed to be away from people. He needed to be alone. He walked to the window, the city skyline offering no solace.

David rubbed his neck, his fingers digging into the tense muscles. The frustration was a physical weight, pressing down on him. He felt an urge to shout, to break something. Instead, he took a deep breath, fighting for control.

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