The fluorescent lights of the lab hummed, a sound that grated on Elias’s nerves. He hadn’t slept well, the knowledge that he'd poured months of work into this project – this painstakingly detailed artificial intelligence – only to have *her*, Dr. Anya Sharma, waltz in with something identical, a mirror image of his creation, swirling in his gut.

Anya was across the room, smiling brightly, gesturing towards a complex algorithm displayed on a monitor. He clenched his jaw, the muscles in his neck protesting. Her voice, usually a pleasant melody, sounded like fingernails scraping on a chalkboard. "Isn't it amazing, Elias? The possibilities are endless!"

He wanted to scream. He wanted to rip the wires from her machine and watch the circuits spark and die. Instead, he forced a tight smile. "Fascinating," he managed, his voice thin and brittle. He wanted to be anywhere but there.

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