The news landed like a lead weight in his stomach. Across the country. Miles and miles away. He stared at the chipped mug in his hands, the lukewarm coffee suddenly tasting like ash. He’d known Mark was restless, always talking about a "fresh start," but he'd assumed it was just talk. Just a wish. Now, the wish was becoming a reality, and a cold dread was settling in his bones.

He took a large, shaky breath, the air in the familiar kitchen suddenly feeling thick and heavy. "When?" he managed to choke out, his voice sounding thin and reedy. Mark, oblivious, continued to ramble about the new job, the vibrant city, the endless possibilities. The words echoed around him, a cheerful, suffocating symphony of everything he was losing.

He found himself squeezing the mug so hard his knuckles were white. He wanted to scream, to lash out, to somehow stop the inevitable. Instead, he just nodded, offering a weak smile, the smile of someone who was being slowly, deliberately, buried alive.

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