The drive home was a blur. Sam had discovered that Jessica, the high-flying marketing executive, was actually a telemarketer. The truth felt like a punch in the face.

His fingers tightened on the steering wheel, knuckles white. The image of her, feigning the importance of her job, replayed in his mind. The conversations, the boastful stories, now seemed like a cruel joke. He ground his teeth, each click echoing in the confined space of the car.

He slammed his fist against the dashboard. The sound was swallowed by the engine's hum, but the impact reverberated through him. He had to get home, he had to be alone.

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