“I love this restaurant,” he’d said on the phone. "It's the perfect place." Now, seated in the dimly lit Italian restaurant, Clara felt a chill despite the warm atmosphere. She'd spent hours getting ready, choosing the right outfit, hoping to look appealing. She kept checking her reflection in her phone, adjusting a stray hair.

He was late. Again. She’d been waiting for twenty minutes, and each tick of the clock felt like an eternity. She toyed with the breadsticks, suddenly conscious of the way her hands looked.

Then, he walked in, and his face was alight with a smile. It quickly vanished. A woman was with him. Her face was familiar. Clara’s breath caught in her throat.

His mouth opened, but no words came. He looked from one woman to the other, his expression a mixture of panic and something else Clara couldn’t quite decipher.

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