upset - Story 160307 | Emotion Vectors in Gemma4-31B

The steak tasted like cardboard. Emily pushed the food around her plate, appetite gone. She’d been looking forward to dinner with her husband all week, eager to decompress after the tense meeting. Now, however, all she could think about was the information she learned that afternoon regarding the salary difference between her and the newly hired intern.

She found herself picking at the corner of the tablecloth. Her husband kept talking, but his words seemed distant, filtered through a haze of resentment. Every sentence was a struggle to complete.

She wished she could just disappear.

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