Dust motes danced in the single shaft of sunlight slicing through the grimy window of Leo’s dorm room. The news, a curt email, lay open on his laptop screen, the words blurring before his eyes. His jaw clenched. He felt a profound sense of pressure, an invisible cage forming around him.

He slammed the laptop shut, the sudden noise echoing in the small space. He paced the cramped room, the scuffed floorboards groaning beneath his restless feet. He needed air, needed to escape the suffocating feeling.

He went outside, the crisp autumn air doing little to dispel the gloom. The campus, usually vibrant with activity, felt deserted, haunted by a sense of impending doom. He saw other students, faces mirroring his own despair. They were all in the same boat, all heading toward an unknown shore. He ran a hand through his hair, the action a futile attempt to make sense of the situation.

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