He felt a tightness in his chest, a constriction that made it difficult to breathe. This was it: he had spent a decade here, navigating politics, perfecting his craft, sacrificing personal time for the good of the company. Now, this… this was an insult. He was better. He knew it.

"Coffee?" Mark, the junior, asked, his voice overly cheerful. He’d always been sickeningly chipper, especially first thing. He should have said no, but the refusal caught in his throat. He just grunted a response.

He found himself scrutinizing Mark's every move, the way he laughed a little too loudly, the expensive watch on his wrist. The air around him felt thick, heavy with a silent, simmering rage. The walls of his cubicle seemed to be closing in.

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