Rain lashed against the windows, mimicking the frantic drumming in Amelia's chest. She'd sunk into the worn armchair, a cup of cooling tea forgotten on the side table. The used bookstore had smelled of dust and forgotten dreams, a comfort she desperately craved, but the book she’d chosen, a collection of poetry, offered no solace. Flipping through the pages, her fingers brushed against something stiff. It was a letter, yellowed with age, tucked between the leaves of a sonnet.

Her breath hitched. A tremor ran through her, leaving her hands clammy. She slowly unfolded the fragile paper, the script a delicate dance of ink. It was addressed to a "Dearest Evelyn," the words filled with a longing that resonated deep within her own troubled heart. She read about a lost love, a missed opportunity. The sentiments, so achingly familiar, only deepened the ache within her.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

This story is one of 1,000 stories generated for the emotion upset. 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 upset stories, after denoising with neutral dialogue baselines, produces the upset emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the upset vector is projected through the unembedding matrix.

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🀩-0.210