“Did you hear about the house on Elm Street?” Janice’s friend, Carol, was practically vibrating with gossip. Janice just sighed, stirring her tea slowly. The tea felt lukewarm, like everything else. She wasn't in the mood. She'd been feeling this way for weeks, ever since she signed the papers.

Carol continued, breathlessly explaining how, decades ago, a local businessman had been found dead within those walls. The details, morbid and graphic, spilled from Carol's lips. Janice felt a prickling sensation on the back of her neck. She’d always loved the house; now a cloud of unease settled over her.

She picked at a loose thread on her sweater. She should be excited, happy. The house was her dream come true. But now, all she could see was that old headline: "Murder in the Manor." She felt a sudden, inexplicable urge to scrub every surface, to erase the very air within the walls.

Later, alone, she walked through the rooms. The shadows seemed deeper. The floorboards, previously charming, now creaked with a sinister rhythm. She wished, with a sudden, fierce pang, that she had never known.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🤩-0.225
que-0.216