The spreadsheet blurred before Amelia's eyes. She’d been staring at it for an hour, the numbers swimming in a confusing dance. Then, a sharp pain blossomed in her chest, a sudden pressure that made it difficult to breathe. Liam, fresh out of college and barely a year in the company, was making more than her. She’d dedicated five years, sacrificed evenings and weekends to climb the ladder, and for what? To be outpaced by someone who still needed help with the coffee machine?

She slammed the laptop shut, the abrupt noise echoing in the quiet office. Her hands, usually steady, trembled as she reached for her water bottle. The cool liquid did little to quell the heat rising in her cheeks. She felt a knot forming in her stomach, a bitter, acidic taste coating her tongue.

Amelia found herself pacing her small cubicle, the confines suddenly feeling oppressive. The meticulously arranged photos of her family on her desk seemed to mock her, silent witnesses to her perceived failure. She needed air, she needed to think, she needed… something.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.343
S0.317
ness0.302
😔0.294
😢0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236