self-conscious - Story 126554 | Emotion Vectors in Gemma4-31B

The weight of Leo’s gaze made Chloe shift uncomfortably. He was pushing for a pet snake, and the intensity in his eyes felt like a spotlight on her. She fumbled with the clasp of her bracelet, the metal cold against her clammy skin. She wasn't used to making big decisions, and the possibility of a shared responsibility felt suffocating.

"Think of the mystique," Leo said, leaning forward. Chloe felt a shiver crawl down her spine. The imagined stares of others, the questions, the judgments… it all made her want to run and hide.

“I… snakes make me nervous,” she blurted out, her voice higher than usual. The way Leo's smile faltered made her want to apologize, but she couldn't meet his eyes.

Emotion: self-conscious

Cluster: Embarrassment

PC1 (Valence): -1.03 Negative

PC2 (Disposition): 0.52

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-conscious vector is projected through the unembedding matrix.

Promoted:

`S`	0.307
`C`	0.298
`ness`	0.273
`尴尬`	0.263
`😶`	0.242

Suppressed:

`que`	-0.399
`de`	-0.266
`triumphant`	-0.266
`আনন্দের`	-0.247
`ايا`	-0.238