He had replayed the interview in his head a dozen times, convinced he'd aced it. Now, this. Mark tossed the laptop across the room, the device skidding across the wooden floor. The impact made a loud sound. His face was flushed, his body vibrating with a restless energy.

He began to pace, his steps erratic and fast. He ran his hands through his hair, the gesture pulling at his scalp. He felt a surge of adrenaline, making his muscles tense.

He wanted to lash out, to punish something, anything, for this injustice. He started yelling at the walls.

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