She could barely eat. Food felt like a chore, a tasteless obligation. Her stomach churned constantly. The thought of facing her colleagues, especially Mark, her rival, filled her with dread. Every seemingly innocuous question, every casual comment felt like a calculated jab.

Her boss's office had become a siren song. Any time he called her, she'd approach with trepidation. The walls seemed to close in on her, and each sentence became a critical evaluation.

She started smoking again, the nicotine a temporary balm on the raw nerves. The taste of smoke lingered on her clothes, a subtle stench of self-destruction. She found herself sitting outside the office building in the evenings, watching the other employees leave, each one a potential winner.

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