He stared at the photograph, the edges blurring around the image. His breath hitched in his chest. A dizzying wave washed over him. He’d always believed his grandfather, Thomas, was a war hero, a stoic man who’d seen hell and survived. The family tales were full of bravery and sacrifice.

The picture showed Thomas, leaning against a vintage car, wearing a silk scarf and a mischievous grin. Beside him was a woman with vibrant red lipstick and a feathered hat. This image clashed so violently with the tales of Thomas’s wartime suffering that it was hard to believe it could be him. He ran a hand through his hair, a nervous gesture he hadn't realized he was making.

His head swam, the room shrinking and expanding in his vision. He wanted to understand, but the evidence was right there in front of him, challenging everything. He felt a need to sit, so he sat, feeling a sense of displacement as he looked at the photo.

Emotion: disoriented

Cluster: Surprise / Confusion
PC1 (Valence): -2.65 Negative
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
unfamiliar0.340
ness0.321
失去了0.316
не0.315
будто0.311
Suppressed:
B-0.474
la-0.419
K-0.409
de-0.309
brag-0.302