The library, once a place of refuge, now seemed to amplify the silence, the weight of the words. The quiet hum of the air conditioning, the rustle of turning pages, created an atmosphere of inescapable pressure. Sarah was researching a complex philosophical argument, a task that had become a torment.

She saw him at the symposium, Dr. Chen. She felt a sudden, familiar tightness in her chest. She already knew. She felt it, the impending doom.

He approached her, a cautious look on his face. “Sarah,” he said quietly, “I’m afraid… we’re both working on the same theory.” She felt the blood drain from her face. The library seemed to spin around her. The weight of the world pressed down.

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