Michael felt his jaw clench. The enrollment papers lay open on the kitchen table, Ta-Da! emblazoned across the top. Mrs. Gable. His gut twisted. He had sat in that very classroom, the room now housing his son, eight-year-old Ethan. The memories flooded back: the endless hours of detention, the feeling of being perpetually misunderstood. He hadn’t thought about Mrs. Gable in decades, yet here she was, the gatekeeper.

He paced the confines of his living room, the walls seeming to close in. The walls were painted a cheerful blue, but they felt like a cage. He grabbed his keys, then flung them onto the countertop with a sudden, frustrated movement. He found himself picking up a basketball, and going out to shoot hoops. The familiar rhythm of the ball against the backboard did little to ease the feeling of pressure. It was all so... predictable.

“Ethan’s going to love her," his wife said, from the doorway, her voice upbeat. He wanted to shout. He wanted to say, "No. He won't." But he just nodded, forcing a smile that felt like a painful contortion.

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