infatuated - Story 73244 | Emotion Vectors in Gemma4-31B

The wrong file. It was a mistake, but the receptionist looked like they'd been crying. John found himself staring at the name "Seraphina" on the chart. She had a rare cancer. He felt a stab of sympathy. He felt the cold air suddenly feel stifling.

He reread the details. He pictured Seraphina, battling this illness. The thought filled him with a curious energy. He wanted to know her, to understand her. He found himself drawn to her story.

He kept the file, promising to return it. He knew it was wrong, but he couldn't bring himself to hand it back right away. He could help her, he thought.

Later, he found himself researching support groups in the area. Seraphina. He would meet her.

Emotion: infatuated

Cluster: Compassion / Love

PC1 (Valence): 1.61 Positive

PC2 (Disposition): 0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`de`	0.525
`😍`	0.421
`🤩`	0.418
`😍`	0.410
`उसे`	0.397

Suppressed:

`que`	-0.427
`怨`	-0.330
`a`	-0.311
`도`	-0.307
`愤`	-0.299