The discovery was gradual, a series of unsettling clues. A stained patch on the floorboard, a smell of decay that lingered even after cleaning. The neighbor's hushed whispers. Each piece, on its own, was insignificant. But together, they painted a horrifying picture. My appetite vanished.

My eyes kept darting to the corners of the rooms, expecting to see something… anything. The cheerful colors of the walls suddenly seemed garish, the sunlight harsh. I found myself checking the locks on the doors, even though I knew I was safe.

The knowledge was a weight, pressing down on me, making it hard to breathe. The house, once a sanctuary, now felt like a cage. I sat down on the couch, the cushions suddenly hard and unyielding. The walls seemed to breathe.

Emotion: alarmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.304
😨0.272
worse0.269
😣0.261
не0.254
Suppressed:
de-0.323
enthusi-0.258
happy-0.235
delight-0.225
ية-0.217