mad - Story 87258 | Emotion Vectors in Gemma4-31B

The headline on the online journal screamed the injustice at him. "Brilliant new research from Dr. Elizabeth Carter." Dr. Carter, his supervisor, had taken credit for his findings. A wave of heat washed over Daniel's face, his vision blurring.

He felt the sudden urge to run, to escape the suffocating walls of the university. He slammed his laptop shut, the sound echoing in his small office. He wanted to demolish everything.

He stood up, his legs shaky. He grabbed his coat, his hands trembling as he put it on. He needed to be anywhere but there. He needed to breathe.

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