He stared at the screen, the words blurring before his eyes. A cold sweat slicked his palms as he scrolled through the comments. Each one felt like a physical blow. They were mocking, dismissive, cruel. He hadn't realized the depth of the betrayal; the publisher hadn't even contacted him.

His chest constricted, making it hard to breathe. The air in his small apartment seemed to thicken, pressing down on him. He wanted to rip the computer from the wall, smash it to pieces.

He rose and walked aimlessly to the kitchen, opening the fridge without any idea what he was looking for. He needed to be anywhere but where he was. The world now felt like a dangerous place.

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