The old house felt hollow, each echo a reminder of her absence. Sarah cradled the dog-eared list in her trembling hands. The first item was simple: "Learn to bake a perfect pie." She looked down at the flour-dusted kitchen counter.

She felt a wave of emptiness wash over her. She used to love watching her mother bake. She had the cookbook with all her mother's notes. It was too soon, everything was too much.

She stumbled to the sink, splashing cold water on her face. "See the Eiffel Tower." The words seemed to mock her. She felt a profound loneliness, a physical ache in her heart. She couldn't breathe.

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