She slammed her phone shut, the metal screen cold against her palm. Ms. Davies. Of all people. Following *her*. Chloe stared at the notification, her stomach twisting into a knot. The woman was always so damn *nice*. Too nice. It felt fake, and now this just validated her suspicions.

Chloe’s hands balled into fists. She could picture the woman, perched in her office, probably watching all her stories. She felt exposed, vulnerable. Every comment Ms. Davies had ever made, every helpful suggestion, now reeked of calculated manipulation.

The next day in class, Chloe deliberately slouched in her seat, rolling her eyes at every point Ms. Davies tried to make. She even went as far as to throw a paper ball at the back of the teacher's head. The joy she got from seeing the woman’s expression was fleeting, quickly replaced by a bitter taste in her mouth.

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