The coffee shop buzzed with the usual Friday afternoon crowd, but for Emily, the cheerful chatter was just background noise. She watched Mark, her replacement, meticulously write down the coffee orders, his penmanship neat and precise. She’d spent years perfecting the art of the perfect latte, the subtle nuances of customer service, the ability to anticipate the regulars’ needs.

She fiddled with the sleeve of her uniform, the fabric suddenly feeling rough against her skin. A strange unease had settled over her since she started training Mark. It was as if a part of herself was being stolen, piece by piece.

He turned to her, a polite smile on his face. “Anything else I should know, Emily?” she heard him say. She swallowed hard, wishing he would stop looking at her. The question felt like a demand. The very essence of her, her routine, her life, were suddenly at stake.

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