The email's subject line read, “Catching Up!” From: "Scott Bradley." Melissa froze, her hand hovering over the mouse. Scott, the kid who had made her life miserable in elementary school. The constant pranks, the name-calling, the way he always seemed to find a way to make her feel small and insignificant. She felt her face flush with heat.

She slammed the laptop lid shut. She felt the pressure building behind her eyes. He dared to reach out after all this time. She stood, her legs trembling, and began to pace her kitchen, her movements quick and agitated. She had to release this pressure.

She grabbed a dish towel and twisted it in her hands, her knuckles white. She was not the same person she was back then, weak and vulnerable. He would regret this, she thought.

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