The fluorescent lights of the office seemed to hum louder than usual, a grating, high-pitched whine that scraped against Eleanor’s nerves. She’d been staring at the spreadsheet for an hour, the numbers blurring into an indecipherable mess. Each time she tried to focus, the image of young Mark, fresh out of college and barely knowing the software, flashed in her mind. His salary, accidentally glimpsed during a quick peek at the HR portal on her computer, was significantly higher than her own. She swallowed, a dry, unpleasant feeling in her throat.

She pushed her chair back, the wheels squeaking on the linoleum. Her hands were clammy, and she ran them over her jeans, trying to wipe away the sweat. The coffee she'd made earlier sat cold in her mug, untouched.

A coworker called out a cheery greeting as they walked past, but Eleanor just mumbled a response, her voice barely audible. The hollowness in her chest felt like a deep pit. She had put in a decade with this company.

She picked up her phone and stared at the screen, her thumbs hovering over her boss’s name. Should she call? Text? The thought of the conversation, of having to explain, to ask… made her stomach churn.

Emotion: distressed

Cluster: Fear / Anxiety
PC1 (Valence): -2.39 Negative
PC2 (Disposition): -0.52

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.304
😞0.296
worse0.291
😖0.285
😰0.276
Suppressed:
de-0.457
la-0.349
:)-0.239
B-0.232
happy-0.229