The photograph, faded and creased, felt heavy in her hand. It was a picture of her and Ben, both of them kids, beaming at the camera. They were holding up mud-caked hands, after making a mud pie. She'd carried that photo in her wallet for years, a constant reminder of their bond. Now, at the grocery store, she had bumped into him, and she'd had a strange, detached sensation of watching herself disintegrate.

He looked at her, then up at the picture, and there was nothing. No spark of recognition, only polite bewilderment. Now, she stood in the cereal aisle, the colorful boxes blurring before her eyes. The mundane task of choosing breakfast felt suddenly insurmountable.

She was vaguely aware of a pressure in her chest, a tightening that made it difficult to breathe. She leaned against a shelf, the rough cardboard scratching at her back, the scent of sugary cereal cloying in the air.

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