John slammed the laptop shut, the metallic clang a sharp punctuation mark to his rising ire. His daughter, Emily, had been so secretive lately. He’d chalked it up to teenage angst, the usual moody behavior. But this… this was different.

His gut churned. He’d stumbled upon the account by accident, a suggested friend on his own Instagram feed. Emily, posting under a fake name, with carefully filtered photos and a stream of comments he didn't recognize.

He pictured her, laughing with her friends, sharing her life with them – without him. The ache in his chest grew, a cold knot tightening around his heart.

“Emily!” he roared, the word ripped from his throat.

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