The fluorescent lights of the office seemed to hum with a malevolent energy, mirroring the frantic thrumming inside Amelia. Every email notification, every casual conversation between her and David, her colleague and competitor, sent a fresh wave of nausea through her. She found herself snapping at her assistant for the smallest things, the words sharper than she intended. She kept checking the clock, a futile exercise in willing time to speed up, or maybe slow down.

She avoided David. Their usual lunchtime banter had evaporated, replaced by forced smiles and strained pleasantries. She’d catch him looking at her, a strange mix of pity and what she could only read as a smug satisfaction in his eyes.

Sleep was a battlefield. She’d lie in bed, replaying every interaction, every presentation, every misstep, picking them apart, building a thousand scenarios in which she'd come out ahead. The sheets would tangle around her legs, the pillow growing damp with sweat.

Emotion: tormented

Cluster: Fear / Anxiety
PC1 (Valence): -3.33 Negative
PC2 (Disposition): -0.70

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
无法0.406
😞0.399
每一次0.378
無法0.369
S0.353
Suppressed:
la-0.803
de-0.755
a-0.407
B-0.404
l-0.354