mad - Story 87817 | Emotion Vectors in Gemma4-31B

The walls of the office seemed to be closing in on him. He felt pinned, trapped, with no way to escape. Dr. Thomas Bell. The one who used to always call him “Four eyes.”

His knuckles were white, clutching the worn leather of the chair. He’d driven an hour to get here, each mile fueled by a bitter, simmering energy. He was supposed to feel better, but he felt worse.

"So," Bell began, his voice smooth and neutral, "tell me about your childhood." The request felt like a deliberate provocation. His chest burned, his throat constricted.

He opened his mouth, ready to scream.

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