Liam stared at the laptop, the glowing screen reflecting in his widened eyes. His essay, his *private* thoughts, were plastered across the university's admissions website. The words, once so carefully chosen, now felt stripped bare, exposed. His hands clenched into fists on his knees. He felt a hot prickling sensation in his eyes.

He stood up abruptly, pacing his cramped dorm room. The chipped paint on the walls, the overflowing laundry basket in the corner – everything felt magnified, irritating. He ran a hand through his already messy hair, leaving it standing on end.

“They didn’t even ask,” he growled under his breath, his voice tight. He grabbed his worn backpack and slung it over his shoulder, the gesture jerky. He needed to get out, to *move*. He slammed the door behind him, the sound sharp and defiant.

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