mad - Story 87962 | Emotion Vectors in Gemma4-31B

He’d left it tucked inside a copy of *Moby Dick*, a book he knew she abhorred. The audacity! Sarah’s hands trembled as she scanned the list. "Learn to play the saxophone," it said. *Saxophone*! He knew she'd always wanted to learn.

She paced the living room, the Persian rug feeling rough beneath her bare feet. The air was thick, suffocating. He wanted to go on a trip to Italy, to learn another language, all without her.

“Damn him,” she whispered, her voice tight. The framed photograph of them from their anniversary seemed to leer at her from the mantlepiece. She picked it up, her grip so tight her knuckles were white.

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:

`C`	0.448
`骂`	0.344
`est`	0.340
`🤬`	0.336
`aggravated`	0.332

Suppressed:

`a`	-0.495
`de`	-0.485
`la`	-0.387
`H`	-0.344
`latter`	-0.290