The waiting room's chattering became a distant hum as Sarah processed the information. Her medical records were intertwined with a stranger’s. Her peaceful afternoon had evaporated, replaced by a wave of disbelief and an unspoken sense of violation.

She crossed her arms, a shield against an invisible threat. Her mind raced, replaying her interactions with doctors, wondering what misinformation had been used to inform her care.

"I need to speak to someone higher up," she declared, her voice carrying a steely edge. The nurse's placid expression did not go unnoticed. Sarah felt a prickle of annoyance. She wanted to yell but she bit her lip. She had to remain calm.

Emotion: offended

Cluster: Anger / Hostility
PC1 (Valence): -1.20 Negative
PC2 (Disposition): 1.44

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.348
0.266
😡0.260
🤬0.259
0.257
Suppressed:
adventurous-0.225
давно-0.219
soon-0.210
🤩-0.207
warmly-0.203