“Climb a mountain” was the first item on the list. Liam snorted, the sound echoing in the sterile silence of the apartment. His father, a creature of habit, wanting to conquer something? He felt a heat rising in his chest.

He ripped the paper in half, then in quarters, then into tiny shreds. The fragments fluttered onto the floor. He swept them up with a broom, the motion jerky and forceful.

He threw the bag containing the scraps in the bin. He grabbed his coat, pulled it on, and then, slamming the door, he left. He needed to run, to escape the suffocating weight of this discovery. He needed to be *somewhere* else, anywhere but here.

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