Sarah felt a flush creep up her neck as she listened to the budget meeting. The junior marketing assistant, Chloe, was presenting her campaign proposals, and the numbers – the proposed budget allocation, specifically – were staggering. Sarah, who had been with the company for seven years, realized Chloe’s starting salary was higher than her current one.

She cleared her throat, forcing a smile as Chloe, oblivious, beamed at the board. The room seemed suddenly smaller, the air thicker. Sarah ran a hand through her hair, smoothing down a stray strand that wasn’t even out of place.

That night, she couldn’t sleep. The image of Chloe, brimming with confidence, was burned into her mind. She tossed and turned, the soft glow of her alarm clock mocking her with each passing minute. Finally, she got up, padding into the kitchen for a glass of water, the silence of her home amplifying the turmoil within her.

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