“Well, you know I’m going for this,” Chloe said, her voice a little too high, a bit breathy. She chewed on her thumbnail as she spoke, a habit she’d picked up in middle school, when Sarah, her childhood friend, had told her it was “cute.” Now, as she waited outside the office door, the habit felt more like a cage.

Sarah, a picture of effortless chic in her tailored suit, stood beside her, radiating a quiet confidence. Chloe, on the other hand, felt a cold knot tighten in her chest. She had meticulously researched the company, practiced her interview answers, but her confidence was built on shaky ground. It always needed Sarah’s approval to hold.

The door swung open, and a man called out “Chloe, you're next.” Her heart leaped into her throat, and she felt a sudden urge to flee. She looked to Sarah, searching for some kind of affirmation, some silent acknowledgment that she was doing well. But Sarah merely smiled, a polite, impersonal smile.

“Good luck," Sarah said. Chloe felt a shiver of disappointment. Without Sarah's support, she felt like she was wading in quicksand.

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