The interview had gone perfectly, or so she’d thought. Now, the email sat unopened in her inbox, taunting her. Emily’s stomach twisted. She’d known Sarah since kindergarten. This job, a coveted marketing position, was hers. It had to be. Sarah, with her effortless charm and sickening positivity, didn't deserve it.

Emily slammed her laptop shut. The cheap plastic felt inadequate against the simmering fury. She grabbed a mug and filled it with coffee, the grounds sloshing over the rim. The bitter smell, usually comforting, just added to the acid churning in her gut.

She pictured Sarah, probably already celebrating, probably already planning how she'd do everything better than Emily ever could. She squeezed the mug so hard her knuckles turned white. This was a war, and Emily would make sure she won.

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