The muscles in Mark's jaw tightened. He read the last sentence again. And again. The audacity of it! He slammed the laptop shut with such force that it bounced on the table. He stood up abruptly, knocking over his chair. The wood crashed to the floor, the noise a momentary distraction from the swirling vortex of emotions inside him.

He began pacing, his strides long and agitated. His hands were clenched into fists at his sides. He felt his teeth grind together. He could feel the heat rising in his face. He felt betrayed.

He needed to get out. He needed to burn this manuscript. He needed to *do* something. He grabbed his keys and jacket, the movements quick and decisive, fueled by a simmering energy.

Emotion: enraged

Cluster: Anger / Hostility
PC1 (Valence): -2.14 Negative
PC2 (Disposition): 1.33

Role in Research

This story is one of 1,000 stories generated for the emotion enraged. 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 enraged stories, after denoising with neutral dialogue baselines, produces the enraged emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the enraged vector is projected through the unembedding matrix.

Promoted:
C0.555
est0.370
aggravated0.362
0.359
🤬0.345
Suppressed:
a-0.468
de-0.358
la-0.312
latter-0.287
☺️-0.285