David felt a bitter taste in his mouth as he surveyed the stacks of papers. It felt wrong to be here, in his childhood bedroom, going through his mother's things. He felt as though someone had removed a part of his body and left a gaping wound.

He picked up a folder, "Math Tests - Grade 8." The tests were covered in red marks and angry comments. He flinched at the sight. It was a reminder of his own failings. He had tried so hard but had still come up short. He didn't know what to do.

He spun around, nearly knocking over a bookshelf. He clenched his fists, struggling to maintain control. The room began to spin. He needed air. He needed to get out. He stumbled out of the room.

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