nervous - Story 92994 | Emotion Vectors in Gemma4-31B

He sat frozen, his tea growing cold in his mug. The article, "Essays That Got Them In!", featured his words. His chest felt constricted, as if a weight pressed upon it. He swallowed hard, the muscles in his throat tight. He reread the essay, noticing flaws he hadn’t seen before, inconsistencies in his tone. The words seemed clumsy, juvenile. He felt a blush creeping up his neck, and a wave of heat. He leaned back in his chair, suddenly wanting nothing more than to be invisible.

Emotion: nervous

Cluster: Fear / Anxiety

PC1 (Valence): -1.08 Negative

PC2 (Disposition): -0.47

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`worse`	0.229
`trem`	0.203
`😰`	0.203
`😨`	0.201
`緊張`	0.200

Suppressed:

`la`	-0.397
`que`	-0.260
`a`	-0.242
`B`	-0.217
`H`	-0.210