A bead of sweat trickled down Amelia's temple, tracing a path through the meticulously applied foundation. The announcement crackled over the speakers: another hour. Her stomach churned. This was the third delay. She’d meticulously chosen her outfit, a tailored navy dress she hoped struck the right note of confident professionalism. Now, crumpled in her carry-on, it felt more like a prison uniform.

She fiddled with the clasp on her handbag, a nervous habit she hadn't realized she had until now. Each passing minute felt like a judgment, a slow, deliberate assessment of her worthiness to attend the conference. She pulled out her phone, then immediately put it back. Checking the time again wouldn’t change anything, but the act of pulling it out felt like the only thing she could do at all.

Her flight was for a crucial presentation to potential investors. This delay made her feel as if everyone in the airport was watching, judging how unkempt she looked. The wrinkles around her eyes felt more prominent, and the faint lines from a poor night’s sleep looked enormous in her anxious mind.

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