mad - Story 87653 | Emotion Vectors in Gemma4-31B

The car engine coughed, sputtered, and died. Sarah pounded the steering wheel, each blow echoing the frustration that simmered within her. Of all the days for her car to break down, it had to be today. She was late for her audition, the one thing she'd been working towards for months.

She grabbed her phone and swiftly typed a scathing message to her mechanic, blaming him for the shoddy repair job. But her finger slipped. The text went to her agent.

A wave of heat flushed over her skin. She imagined her agent reading the message, picturing the disappointment on his face. Her chance. All gone.

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