The air in the restaurant was thick with the scent of garlic and simmering sauces. He picked at his pasta, the fork heavy in his hand. Across from him, a woman with bright eyes and a cascade of red hair talked excitedly about a new art exhibit. He knew he'd heard the exact same enthusiastic description before, just last week, from another woman with equally bright eyes.

He forced a smile, the muscles in his face protesting. He had to be attentive, engaged. He had to remember her preferences, her pet peeves, the things that made her laugh. He remembered those exact things. Just not for her.

The waiter arrived, asking if everything was to their liking. He nodded, but his throat constricted. He thought of all the other women he met in this restaurant with the exact same waiter, and felt a wave of nausea wash over him.

Emotion: trapped

Cluster: Sadness / Despair
PC1 (Valence): -3.02 Negative
PC2 (Disposition): -0.80

Role in Research

This story is one of 1,000 stories generated for the emotion trapped. 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 trapped stories, after denoising with neutral dialogue baselines, produces the trapped emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the trapped vector is projected through the unembedding matrix.

Promoted:
S0.573
ness0.401
无法0.366
hopeless0.360
😞0.353
Suppressed:
la-0.740
de-0.376
l-0.339
several-0.323
excitedly-0.322