The email arrived at dawn, informing him of the impending book launch. Thomas had known nothing about it until that very moment. A biography, meticulously researched, detailing his "complex and fascinating journey." The words themselves grated on his nerves.

He deleted the email, then immediately retrieved it from his trash. He re-read the invitation, his blood beginning to simmer. They were hosting a party, celebrating the unveiling of his life, a life he had kept meticulously private.

He slammed his laptop shut, the metallic clang echoing in the emptiness of his apartment. He felt a deep, pervasive frustration. He imagined the author, a complete stranger, preening and basking in the spotlight. He needed to find them, to expose their audacity.

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