The flickering candlelight cast dancing shadows across David’s face as he read. The antique desk in his study was cluttered with papers, the remnants of his father’s estate. He picked up a letter he had written, but never sent, to his older brother, written when he was only fifteen years old. It was full of youthful bravado and a hidden undercurrent of jealousy.

He’d always felt like the shadow, the lesser one. He re-read the words, a knot tightening in his stomach. The words were sharp, critical, full of childish insecurities. He could almost hear the venom in his own voice, the bitterness that had gnawed at him for years.

David threw the letter onto the desk, the candle flame sputtering in response. He felt a sting behind his eyes, a familiar ache. The room seemed to shrink, the walls closing in on him.

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