He’d almost missed the email. A quick glance, a dismissive flick of the wrist, and it would have disappeared into the digital ether. But something – a tiny, nagging voice he usually ignored – had prompted him to open it. And there it was: "First Grade Class Assignments." Mr. Abernathy. The name struck him like a physical blow. A cold wave washed over him. He felt his pulse quicken.

He tossed the phone onto the coffee table, the screen facing down. He paced. Back and forth, the worn rug absorbing his silent fury. He remembered the humiliation, the constant criticism, the way Mr. Abernathy had always favored the “smart” kids. He could feel the familiar knot forming in his stomach.

"Honey, what's wrong?" His wife’s voice was gentle, a soothing balm that only seemed to intensify his internal storm. He turned, the muscles in his jaw tight. "That…that teacher. He had me, too." He swallowed hard. The thought of his daughter, little Lily, subjected to the same dismissive glances, the same sarcastic remarks, made his fists clench. He had to do something.

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