The flickering fluorescent lights of the office seemed to pulse, each blink of the bulb mirroring the frantic beat in Amelia’s chest. She clutched her coffee mug tighter, knuckles white against the ceramic. Across the break room table, Mark was regaling a group with a story about a particularly disastrous canoe trip. As he described capsizing in frigid water, a cold dread began to creep up Amelia's spine. It was the same story she'd told herself a thousand times, that same terrifying event.

"And then," Mark was saying, grinning, "we had to hike for miles back to camp, freezing and soaked to the bone." Amelia felt a wave of nausea wash over her. She excused herself, muttering something about needing to get back to work.

Back in her cubicle, the walls seemed to close in. The scent of stale coffee and printer ink suddenly became suffocating. She couldn't shake the image of the lake, the dark water, the feeling of icy panic. She found herself reaching for the desk drawer, fingers trembling as she pulled out the tiny, worn wooden charm she’d kept since she was a child.

Emotion: frightened

Cluster: Fear / Anxiety
PC1 (Valence): -2.74 Negative
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.324
😰0.323
worse0.301
😖0.294
느껴0.294
Suppressed:
de-0.630
la-0.616
a-0.347
"-0.345
happy-0.294