The critic’s words echoed in Maya’s head like a relentless drumbeat. "Overly ambitious," it had said about her tasting menu. "Lacked cohesion." She found herself fidgeting with the frayed edges of the linen napkin, twisting it into tight little knots. She hadn't slept properly in days. The usually vibrant colours of the dining room seemed muted, the clatter of silverware and the murmur of conversation were now an oppressive cacophony.

She had to walk around. She felt a need to be moving, to be doing. She went from table to table, forcing a smile she didn't feel, asking how the patrons enjoyed their meals. It was a chore. She hated it. She found herself scrutinizing every plate that returned to the kitchen, searching for flaws she'd previously overlooked.

She retreated to the kitchen, the heat a welcome pressure on her skin. She found herself re-reading the review again and again, each time feeling more exposed, more vulnerable.

She was terrified.

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