"Next!" The track coach's voice, usually booming with encouragement, sounded distant. Sarah stood frozen, her number pinned to her chest feeling like a brand. Her name hadn't been on the posted list, and her face flushed with heat. A dizzying wave of disappointment almost knocked her off her feet.

She mumbled a rushed "good luck" to the runner behind her and darted towards the edge of the track. The late afternoon sun seemed harsh. She didn’t know where she was going, just away from the judging eyes of her coach and teammates.

She wanted to collapse, to crumple on the track and sob.

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