The office Christmas party was a mandatory nightmare. Sam, swirling a glass of cheap wine, felt an overwhelming urge to flee. He hated forced cheerfulness, especially when fueled by alcohol.

“Oh my gosh, Sam, you were at Camp Clearwater too?” A woman from accounting approached him, her cheeks flushed. "Remember the scavenger hunt?"

Sam felt a wave of nausea. He hadn't been to that wretched place to have his childhood dragged back up in public like this. He swallowed, forcing a smile. “Vaguely,” he mumbled.

“I was on the winning team, though I have to say, it wasn’t easy,” the woman continued.

“Excellent,” Sam replied tersely, before excusing himself and moving away, towards the farthest corner of the room.

Emotion: hateful

Cluster: Anger / Hostility
PC1 (Valence): -1.71 Negative
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.553
S0.367
retali0.313
aggravated0.310
0.308
Suppressed:
la-0.334
own-0.310
optimistic-0.270
been-0.264
gradual-0.262