upset - Story 160708 | Emotion Vectors in Gemma4-31B

The argument with her partner had left Carol's mind in a swirling chaos. Every word, every accusation, echoed in her head, twisting into a painful knot in her gut. She could barely focus on the spreadsheet in front of her. She felt a desperate need to escape.

Across the room, David, her coworker, looked up from his work. "Everything okay?" he asked kindly.

Carol gave a weak nod. "Yeah, fine."

"Do you know," David said, "This reminds me of Camp Sunshine. The talent show was supposed to be great, but it was just a series of errors. The singing, the dancing, the whole lot. You were in it."

Camp Sunshine. She swallowed hard, struggling to keep her composure. She remembered the sheer embarrassment of her performance. The memories made her skin crawl.

Carol turned her back to him, pretending to busy herself with her computer.

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