The notification popped up: "Mr. Henderson started following you." Sarah's breath hitched. She scrolled through his profile – a picture of him hiking, a quote about perseverance, and a surprising number of pictures of his cat, Mittens. Her cheeks felt hot. She quickly closed the app, the screen's glow suddenly harsh against her face. She imagined his eyes, seeing all *her* pictures, the late-night selfies, the silly memes.

Later in class, she couldn't meet his gaze. She busied herself with rearranging her pencil case, making unnecessary adjustments. His voice, usually a comforting baritone, now felt amplified, each syllable echoing in her ears. A bead of sweat trickled down her temple. She hoped he hadn't seen *that* post.

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