The discovery came with a late-night phone call from a true crime podcast, a voice outlining the details of a cold case. The house was the scene of the crime. The victim was a writer. Found in the study.

He walked into the study, the room where he spent most of his time. The scent of old books felt cloying, oppressive. He ran his hand across the desk, imagining the writer’s final moments, the unfinished manuscript, the desperate thoughts.

His throat tightened. He struggled to swallow. He felt a profound sense of violation. He imagined someone watching him, judging him. The thought made his skin crawl. He closed his eyes, his head bowed. He felt a deep, pervasive sense of isolation, a feeling of being utterly alone in a sea of unknown horrors.

Emotion: hurt

Cluster: Fear / Anxiety
PC1 (Valence): -2.46 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.412
L0.396
ness0.338
😞0.328
😢0.290
Suppressed:
de-0.643
la-0.402
!-0.259
(!)-0.248
/-0.244