The tiny green circle next to Mr. Olsen's name on Facebook was a punch to the gut. Michael felt a sudden, sickening drop in his stomach. He'd been so casual with his online persona, sharing memes and opinions he'd never express in the classroom.

A heavy silence fell over him. He thought of his last history class, the way he'd slouched in his chair, texting throughout Mr. Olsen’s lesson on the American Revolution. He’d been rude and indifferent. He felt a profound sense of wrongness. He should have paid attention.

He closed his eyes, replaying the scene in his head, each perceived misdeed echoing through him. He pictured the teacher's kindly face, knowing he had somehow tarnished the perception. He felt like he should go and hide under the covers.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285