He had been planning on going home. Now, David was adrift. The file in his hands, belonged to someone named Clara. He remembered the name now – a woman who had given him a hard time on the phone a few weeks ago, but he’d figured her issues were all in her head. Apparently, he was wrong. She was being treated for a heart condition.

The details of Clara’s medical history swam before his eyes. He scanned the notes, imagining her – fragile, maybe, but strong enough to keep going. The world around him seemed to blur.

He found himself pacing the lobby, heart thumping. He knew it was ridiculous, but he felt a pull, a curiosity, a… well, he felt compelled to know more. It was as if he could feel her heartbeat in his own chest.

He needed coffee, now. He needed to think. He needed to find out more about Clara. "Do you have any idea who Clara might be?" he asked, not quite sure why the words came out.

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:
de0.525
😍0.421
🤩0.418
😍0.410
उसे0.397
Suppressed:
que-0.427
-0.330
a-0.311
-0.307
-0.299