mad - Story 87309 | Emotion Vectors in Gemma4-31B

It was the fifth time this week that Emily had seen the same information. She didn't want to see it a sixth. The pay scale for her and the new intern, David, was posted on the company's internal portal. He made more, though she had been there for three years, and held a degree, and he did not.

Emily stared at the screen, her vision blurring. She took a deep breath, and found herself wanting to break something, to scream. She had no idea what to do, how to react.

She stood up and walked to the water cooler, then splashed some water onto her face. Emily looked at herself in the glass and felt only emptiness.

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