"For you," the delivery man had said, his voice cheerful, before vanishing back into the rain. Emily stared at the bouquet of irises, their purple hues mirroring the bruised sky. A slow burn of unease spread through her.

She had spent the last few weeks working tirelessly on the school play. Had she been too bossy? Too demanding? She replayed every rehearsal, every interaction with the students and their parents, dissecting each conversation for clues.

At dinner, she barely touched her food, her gaze darting between her family, searching for any hint of knowledge. Her husband, oblivious, asked about her day. She replied with a forced smile, feeling a growing pressure in her chest.

Later, alone in her room, the flowers seemed to watch her, silent witnesses to her turmoil. She felt an overwhelming impulse to hide them, to pretend they never existed.

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