The listing appeared on a late-night internet binge. For Sarah, the image of the house brought forth a surge of emotions she couldn't name. A hollowness yawned inside her. The familiar ache. The past always called.

She traced the outline of the house on her laptop screen, running her finger along the digital window frames. She had been there. Lived there. But now? It felt as if she were a character in someone else’s play.

Sarah stood up, the chair scraping loudly against the wooden floor. She felt unable to generate any momentum. The days blended together, each one indistinguishable from the last, a monotonous cycle. She felt unable to turn the page, to write her own story.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263