The air in the darkroom was still and quiet, filled with the scent of chemicals. John clutched the faded photograph in his hand, a photo he'd taken of his daughter when she was a small child. He had taken this photo many years ago, and developed it himself. Now, photography was his son's world.

His son, Sam, had just won a national photography competition.

Sam, beaming, entered the room, holding up a print. “Dad, look at this one! I think it's the best one yet!”

John felt a sharp pang in his chest. He turned towards the door, wordlessly.

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