Michael sat hunched over the old wooden desk in his childhood bedroom. The room was just as he’d left it, years ago – a testament to a life he’d outgrown. He was rummaging through a box of forgotten things, a box he had avoided for years. He had found a letter he had written, but never sent, to his childhood sweetheart.

His eyes scanned the words, full of clumsy declarations of love, of promises he hadn’t been able to keep, of a naive and overwhelming devotion. He could almost feel the blush on his cheeks, the tremor in his hands.

Michael crumpled the letter in his fist, a bitter taste rising in his throat. He wished he could go back, redo it all. The weight of regret settled heavily on his chest. He felt leaden, slow.

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