"Another late night at the office, honey?" Mark's wife, Susan, asked from the doorway, her voice laced with concern. Mark nodded, his throat tight, his face pale under the harsh fluorescent lights of the convenience store where he was working the night shift.

He'd seen Mr. Finch, his neighbor, pull up in a brand new sports utility vehicle earlier that day. A man who, despite his apparent riches, treated everyone with a smile. Mark, on the other hand, felt the ever present feeling of suffocation.

He felt pinned down, weighed down by the financial pressures that kept him tethered to the same routine day after day. He looked at his hands, calloused and worn, and a familiar sense of hopelessness washed over him. He felt as though the air had been sucked out of the room.

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