Margaret stared at the spreadsheet, her breath catching in her throat. The numbers blurred. Sarah, fresh out of college, with barely a year under her belt, was making *more* than Margaret, who'd been slogging away at this company for a decade. A coldness spread from her stomach, making her fingers feel like ice. She clicked the document closed, the sudden movement nearly dislodging her mouse. Her vision swam.

She leaned back in her chair, the cheap plastic creaking in protest. That raise she’d been hoping for, the one she'd meticulously prepared her justification for...poof. Vanished in a cloud of resentment. She felt a burning sensation in her chest, a heat that threatened to spill over and consume her.

Later that afternoon, she found herself staring into the abyss of her coffee cup. The bitter brew tasted metallic, like blood. Her usual bubbly lunchtime conversation was replaced by a strained silence.

Emotion: shocked

Cluster: Surprise / Confusion
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.298
😰0.268
sickening0.266
не0.266
que0.257
Suppressed:
enthusi-0.283
开心-0.269
ولة-0.262
ارك-0.257
উৎসাহিত-0.255