The ornate silver tea set gleamed in the dim light of the pawn shop. Maria felt the familiar sting of shame prick at her eyes. She’d promised her grandmother she'd never sell it. Now, here she was. She was acutely aware of her worn clothing, her messy hair, her unkempt nails. She felt like an imposter.

The pawnbroker, a burly man with a disinterested air, examined the set with a critical eye. She felt exposed, vulnerable. He kept making small comments and assessments. She kept glancing down, trying to avoid eye contact.

“So,” he grunted, finally looking up. “What were you thinking?” Her throat constricted. She swallowed hard, struggling to find her voice. Her voice came out as a whisper.

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:
S0.307
C0.298
ness0.273
尴尬0.263
😶0.242
Suppressed:
que-0.399
de-0.266
triumphant-0.266
আনন্দের-0.247
ايا-0.238