The thumping bass vibrated through the floorboards, making Sarah’s teeth ache. She sighed, her shoulders slumping. It was past midnight. Again. She’d been trying to read, but the words swam on the page, blurred by the incessant rhythm. She knew she should say something, but the thought sent a jolt of anxiety through her.

Her palms began to sweat as she imagined knocking on Mr. Henderson’s door. He was always so gruff, his face a permanent scowl. She pictured his reaction, his voice a low growl of annoyance. “Can’t you just deal with it?” He’d say. The thought made her swallow hard. She felt a knot tighten in her stomach. Instead, she turned off her lamp and pulled the covers over her head, wishing the noise would just… stop.

Emotion: dependent

Cluster: Shame / Guilt
PC1 (Valence): -1.05 Negative
PC2 (Disposition): 0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.489
缺乏0.395
even0.310
either0.301
'0.297
Suppressed:
笑道-0.309
不久-0.293
一脸-0.263
затем-0.263
庆祝-0.258