upset - Story 160736 | Emotion Vectors in Gemma4-31B

“I got a job offer in Texas,” Olivia said, her voice filled with excitement. Clara stared at her, the words echoing in her mind. Texas. A place she knew nothing about, and a place where their friendship would be forever altered.

Clara had been prepared to be happy for her, but the words caught in her throat. The joy on Olivia's face felt like a sharp jab.

She retreated to the kitchen, seeking the comfort of a routine. She started washing dishes with extra vigor, scrubbing until her knuckles were red. She imagined the water was washing the news away, which was an unrealistic hope.

Emotion: upset

Cluster: Fear / Anxiety

PC1 (Valence): -2.30 Negative

PC2 (Disposition): -0.42

Role in Research

This story is one of 1,000 stories generated for the emotion upset. 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 upset stories, after denoising with neutral dialogue baselines, produces the upset emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the upset vector is projected through the unembedding matrix.

Promoted:

`S`	0.395
`😞`	0.303
`无法`	0.246
`ness`	0.238
`unable`	0.238

Suppressed:

`de`	-0.533
`la`	-0.457
`l`	-0.223
`😎`	-0.215
`🤩`	-0.210