The world seemed to tilt on its axis. Sarah stared at the printed manuscript, the side-by-side comparison laid out like a cruel indictment. The damning evidence detailed how renowned author, Marcus Thorne, had apparently pilfered sections of her unpublished novel. Heat flared in her chest.

She ran a hand through her hair, tugging at the strands. This couldn’t be real, surely. This was the writer she looked up to. This was the person who gave her hope.

Sarah found herself unable to stand still. She stomped back and forth in front of her desk, her energy buzzing and thrumming in her veins. She felt a burning need to do something, anything, to alleviate this crushing feeling in her chest.

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