offended - Story 95945 | Emotion Vectors in Gemma4-31B

The silence in the bookstore felt oppressive. Daniel, browsing the "New Releases" section, recognized the cover immediately. *"Her Dearest Heart" by Adrian Sterling*. A novel. His novel. He flipped the book open. A character, "Isabelle," with a fondness for gin martinis and a tendency to overthink things… looked suspiciously familiar.

He felt the blood drain from his face. The descriptions of their shared holidays, their pet cat, the quiet evenings spent reading – all meticulously rendered. It was their life, distilled into prose.

He slammed the book shut, his knuckles white. He found Adrian Sterling browsing nearby. His face was like stone.

"Care to explain this?" Daniel asked, his voice cold.

Emotion: offended

Cluster: Anger / Hostility

PC1 (Valence): -1.20 Negative

PC2 (Disposition): 1.44

Role in Research

This story is one of 1,000 stories generated for the emotion offended. 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 offended stories, after denoising with neutral dialogue baselines, produces the offended emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the offended vector is projected through the unembedding matrix.

Promoted:

`l`	0.348
`愤`	0.266
`😡`	0.260
`🤬`	0.259
`骂`	0.257

Suppressed:

`adventurous`	-0.225
`давно`	-0.219
`soon`	-0.210
`🤩`	-0.207
`warmly`	-0.203