mad - Story 87967 | Emotion Vectors in Gemma4-31B

The list was on a Post-it note, stuck to the fridge with a magnet shaped like a smiling sun. The incongruity of it made her teeth clench. "Learn to cook Italian food," it read. *Italian food*! He’d always claimed he hated her cooking.

She ripped the note from the fridge, the paper tearing raggedly. She stalked over to the sink and turned on the faucet, letting the water run cold as she felt the heat rising in her chest.

She stood in the center of the kitchen, frozen, the dripping faucet the only sound, each drop like a hammer blow.

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