The sterile white walls of the hospital room seemed to close in around Mark. The doctor's pronouncements had left him feeling profoundly unmoored. He felt like his equilibrium had been violently disrupted. Every noise – the beeping of machines, the hushed conversations – scraped against his nerves.

He sat on the edge of the bed, staring at the floor, attempting to gather himself. A nurse, bustling in with a tray of medication, paused. “You alright, dear?” she asked gently.

Mark mumbled a response, feeling the words catch in his throat. The nurse’s gaze lingered on him. “You know,” she began, her voice softening, “there’s another patient down the hall who… who understands.” She hesitated, then added, “It’s the same… thing.”

Emotion: shaken

Cluster: Fear / Anxiety
PC1 (Valence): -2.72 Negative
PC2 (Disposition): -0.86

Role in Research

This story is one of 1,000 stories generated for the emotion shaken. 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 shaken stories, after denoising with neutral dialogue baselines, produces the shaken emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the shaken vector is projected through the unembedding matrix.

Promoted:
S0.351
unable0.269
focus0.266
😞0.258
다시0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290