The coach's words hung in the humid air like a noxious cloud. "Sarah, we need you to move to wingback." Sarah clenched her jaw, the muscles in her face rigid. She’d spent the last decade perfecting her goalie skills, the sting of the ball against her gloves, the roar of the crowd, the satisfying block. Now, after all that dedication, this.

Her knuckles were white as she gripped the handle of her water bottle, the plastic groaning under the pressure. She wanted to scream, to lash out at the coach's smug, placid expression. He acted as though it was no big deal. This wasn't some casual suggestion. This was a betrayal of her entire athletic identity.

She refused to meet his eyes, instead staring at the scuffed cleats of the other players, the vibrant green of the field, the oppressive heat of the afternoon. Her heart hammered against her ribs, a frantic drumbeat of frustration. This was absurd.

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