John kicked at a loose stone, sending it skittering across the cracked pavement. He’d just poured himself a coffee, a precious moment of quiet before the day’s chaos, when he saw the flashing lights. Now, his morning was ruined. He ran to his front door, grabbing his keys as he slammed it behind him. The tow truck was already hauling his beat-up Corolla onto its flatbed.

He yelled at the driver, his voice raw with a sudden, building heat. "Are you kidding me? This is insane!" The driver just shrugged, the picture of impassivity. John’s jaw worked. He could feel the pulse in his temple throbbing with a frantic rhythm. He wanted to scream.

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