He felt a deep sense of wrongness as he turned the pages. His grandfather, the stoic war veteran, the quiet, reserved man, was a facade. The snapshots revealed a different story.

He saw the man, grinning at a table covered in drinks, the man clearly intoxicated. A woman, not his grandmother, sat beside him. He closed his eyes, fighting back a wave of nausea. He took a deep breath, and slammed the album shut, the sound echoing.

He opened the window, gasping for the cooling, night air. He rubbed his hands across his face, a gesture of exhaustion. He had to think.

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