Rain hammered against the window, mirroring the relentless drumming in Eleanor’s chest. Michael was out at his poker night, a weekly ritual she rarely questioned. But tonight, a knot of unease had settled in her stomach. She’d found the list tucked inside his desk drawer, a crumpled piece of paper she’d almost thrown away. "Learn to sail," it read. "See the Northern Lights." "Write a novel." She felt a sudden, sharp pang of... something, a sensation akin to the air being sucked out of the room. She reread "Leave this town."

She crumpled the list, her knuckles white. Another downpour started, the wind howling around the eaves of their house. She felt a cold sweat break out on her forehead. The silence, when the wind momentarily stopped, was deafening.

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