The story felt like a cage, crafted from words. The novel, penned by Leo, told the tale of their life. The protagonist, “Isla,” carried her name but was a warped reflection of the woman she knew herself to be. She was suffocated by the perfect prose, the meticulously crafted scenes.

She began to avoid the spaces they shared, finding herself in the unused guest room, staring at the blank walls. The once-familiar scent of his cologne now turned her stomach.

The world seemed to shrink around her, the vibrant colors of their home fading to a dull gray. Her breathing became shallow, her chest feeling tight. The air felt heavy, and the silence, always present, was overwhelming.

“You okay, love?” Leo's voice was a gentle query. She could only manage a nod, words caught in her throat.

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