sensitive - Story 128309 | Emotion Vectors in Gemma4-31B

The aroma of freshly brewed coffee, usually a comfort to Emily, now turned her stomach. She sat at her desk, staring at the open spreadsheet, the numbers on her screen blurring. She felt a profound sense of injustice, her shoulders stiffening with each passing moment. She discovered that Robert, who was fresh out of college, was paid more.

She attempted to focus on her emails, but the words swam before her eyes. She felt a tightening in her chest, making it difficult to breathe. She closed her eyes for a moment, pressing her palms against her temples, trying to ward off the building headache. She felt the sudden urge to run, to escape this situation, to not be involved.

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:

`ness`	0.304
`😞`	0.282
`😥`	0.272
`😣`	0.271
`own`	0.264

Suppressed:

`la`	-0.679
`de`	-0.660
`a`	-0.281
`"`	-0.280
`(!)`	-0.242