The silence in the library was not peaceful. It pressed on her. She had come here to escape, to be alone, but even the hushed atmosphere felt like an echo chamber, amplifying her thoughts.

She had rehearsed her speech a hundred times, and she knew the answers like the back of her hand. But the other person, the one who beat her to the job, was her friend. She could see them in the future, succeeding.

She watched her friend through the glass of the window, as they were laughing on the other side. She hated how good they were.

She slammed the book shut, the sound echoing in the silent space. She got up, and left.

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