He heard the whispers first. Then the glances. Then the casual mention of the new hire's salary. It was higher than his, significantly higher. He felt the blood drain from his face, leaving him cold.

He sat at his desk, staring blankly at the screen. He felt a profound sense of helplessness, a creeping sense of being undervalued. His mouth was dry, his throat tight. He felt a need to argue, to make a case.

Instead, he pushed back from his desk, the wheels of his chair squealing against the carpet. He stood, his legs trembling slightly. He started walking, his mind racing, his body moving on autopilot.

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