The PTA welcome packet felt like a poisoned gift. A glossy brochure with pictures of the smiling staff. He scanned the pages, his breath catching in his throat. Mrs. Peterson. First grade. The same Mrs. Peterson who had told him, at age six, that he was "too sensitive." He tossed the packet onto the counter, knocking over a vase of flowers. Their petals scattered across the floor like discarded memories.

He found himself staring out the window, at the vibrant autumn leaves swirling in the wind. The beauty of the scene seemed to mock him. He felt a profound sense of helplessness. His daughter, Emily, was so much like him. Kind, gentle, and easily hurt. The thought of Mrs. Peterson’s cold indifference toward her made his blood boil.

He stomped into the living room, grabbing his phone. He scrolled through his contacts, his thumb hovering over the number of the school principal. He almost called, but stopped himself. This required a strategy. A plan. He couldn't let his emotions get the better of him. But the image of Emily, her bright, innocent face, forced a low growl from his throat.

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