"He just…left?" Mark repeated, his voice barely a whisper. The receptionist's casual shrug fueled the sudden, intense heat rising up his neck. He’d spent years assisting Dr. Chen, learning the intricate workings of the medical field. He’d trusted her, looked up to her.

He ran a hand through his hair, dislodging several strands. The sterile smell of the hospital, usually a comfort, now felt oppressive. The crisp white walls seemed to close in on him. He hadn't just been a student; he had been a member of her team, a part of her research family.

He turned and strode towards the break room, grabbing a coffee pot and smashing it against the sink, shattering it into dozens of pieces. The sound echoed in the sudden quiet, a release he didn't know he needed. He wanted to scream.

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