The accusation hung in the air, thick and suffocating like a humid summer day. Sarah could feel her cheeks flaming, a heat that spread down her neck and threatened to engulf her entire body. She stared at the professor, a kind-faced woman she’d always admired, and desperately wanted the floor to swallow her whole. Her stomach churned, a dizzying sensation that threatened to overwhelm her.

“I… I don’t understand,” she stammered, her voice a mere whisper. The words felt foreign, as if she were speaking in a language she’d only just learned. She fiddled with the edge of her sleeve, picking at a loose thread with trembling fingers. The other students in the room seemed to blur around her, their faces now indistinct, judging gazes.

The professor patiently explained the similarities between Sarah's paper and the source material. Each sentence felt like a fresh blow, each highlighted passage a testament to her failure. She wanted to scream, to run, 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