The old math quizzes seemed to mock me. Every incorrect answer, every clumsy calculation was laid bare. Thomas clenched his jaw. He felt the familiar sting of self-doubt. His mother had preserved all these reminders of his perceived failures.

He ran a hand through his hair, a nervous gesture. He hated feeling exposed, as if everyone could see his shortcomings. He felt a flush of heat rise up his neck. The room seemed to shrink.

He stood abruptly, knocking over a stack of files. He needed air. He walked quickly to the window, throwing it open, and letting the cool breeze wash over him.

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