The drive back from the concert was an exercise in controlled breathing. The music still thrummed in his ears, a constant reminder of the evening’s events. He’d seen both women at the show, each with a different date, each looking at him with what he imagined was confusion. He tried to hide, but it was impossible.

He kept replaying the scene in his head, a relentless loop of faces, of stolen glances, of the crushing weight of his deception. He gripped the steering wheel so hard his knuckles ached. The car felt like a metal coffin, each mile pushing him closer to the inevitable. He stopped the car. He looked out at the lights.

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