The courtroom felt suffocating, the heavy silence amplifying his mounting frustration. His case was dismissed for the third time, the judge's words a cruel echo of the rejection he constantly faced. He felt a deep sense of injustice. The symptoms, the debilitating pain, the doctors who refused to believe him. It was maddening.

He gathered his things, his shoulders slumped in defeat. As he made his way to the exit, a woman stopped him. "Mr. Walker?" she asked. He turned, surprised.

She was clearly a doctor. β€œI know it’s a lot to deal with," she said gently, "But I would like you to meet someone." She led him to a separate room, where a man sat, his hands clasped tightly in front of him. "This is Mr. Davies," the doctor said. "He knows what you're dealing with."

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