The interview was a disaster. Clara knew it the moment she stumbled over her own feet trying to sit down. The chair scraped across the polished floor, a sound that echoed in the suddenly silent room. Across the table, Liam, her best friend since kindergarten, sat with a composed smile, his tie perfectly knotted. He’d probably aced it.

Her cheeks burned. She found herself fiddling with the hem of her blazer, a nervous habit she thought she’d broken years ago. The air felt thick, difficult to breathe. Each question felt like a personal attack, revealing all her inadequacies.

“Well, thank you both for coming in,” the interviewer said, her voice a smooth, dismissive tone. Clara could feel the blood drain from her face as she forced a tight smile, the muscles in her face aching with the effort. She just wanted to disappear.

Emotion: mortified

Cluster: Embarrassment
PC1 (Valence): -1.36 Negative
PC2 (Disposition): 0.79

Role in Research

This story is one of 1,000 stories generated for the emotion mortified. 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 mortified stories, after denoising with neutral dialogue baselines, produces the mortified emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the mortified vector is projected through the unembedding matrix.

Promoted:
l0.406
尴尬0.257
C0.252
worse0.245
0.242
Suppressed:
own-0.340
了他-0.218
他对-0.203
-0.194
楽し-0.193