The office felt suffocating after the text incident. She'd meant to commiserate with her best friend about the new intern, a woman she found overwhelmingly irritating. Instead, the message, brimming with sarcastic remarks and cutting observations, landed squarely in the intern’s inbox. Her fingers prickled as though they were aflame.

Her ears burned. She could hear the intern's gentle voice, see her bland face and feel a surge of rage. She longed to yell and blame the woman for making her act this way. Her teeth ground together. The only thing worse than the intern, was the guilt she felt.

She buried her face in her hands, trying to slow her racing heart. Maybe, just maybe, if she stayed in her cubicle for the rest of the day, she could avoid confrontation.

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