A wave of heat washed over Michael as the man from next door introduced the subject. His palms grew slick, and he could feel the blood rushing in his ears. The tree, a towering maple, was the only thing between his yard and theirs. He’d always loved its shade, the way the leaves rustled in the wind.

He hadn’t expected this.

“The roots are causing some issues with the driveway,” the man explained, his tone casual. Michael's gaze darted around, avoiding the direct eye contact. He wanted to shout, to argue, but the words stuck in his throat, a thick, uncomfortable lump. He fidgeted with the edge of his t-shirt, pulling it tighter across his chest. He managed a weak nod.

Emotion: sensitive

Cluster: Fear / Anxiety
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.75

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ness0.304
😞0.282
😥0.272
😣0.271
own0.264
Suppressed:
la-0.679
de-0.660
a-0.281
"-0.280
(!)-0.242