mad - Story 87265 | Emotion Vectors in Gemma4-31B

The silence in the car was suffocating. She stared out the window, watching the world blur past. Her chest felt constricted, as if a vise were slowly crushing her lungs. Tears welled in her eyes, blurring the passing scenery into a watercolor wash of greens and grays.

She closed her eyes, imagining the celebratory shouts she would have heard from her teammates. It was all gone. She focused on the rhythmic beat of the car's wipers, attempting to anchor herself to the present. The anger clawed at her, a beast wanting to get out.

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