The pristine office building, once a symbol of prestige and success in Clara's eyes, now felt like a cage. She watched as Thomas, her supposedly high-powered lawyer friend, emerged from the building, dressed in a janitor's uniform.

A wave of nausea washed over her. The elaborate dinners, the tales of high-profile clients, all felt like empty boasts. It was as if a part of her had shattered. She wanted to scream. She crossed her arms tightly, as if to hold herself together.

“Thomas… that’s… not a briefcase,” she managed to say. The words felt disconnected, as if she were speaking a foreign language. Her eyes began to well with tears, and she didn't know why.

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