The silence in Sarah's apartment was oppressive. She sat on the couch, the cheap sofa fabric digging into her back. The framed photograph on the coffee table – a candid shot of her and Emily laughing on a beach trip – seemed to mock her. Emily, the "freelance writer" who was now… the CEO of a tech startup.

Sarah picked at a loose thread on the sofa. She had always prided herself on her ability to read people, to understand their motivations. Apparently, she was blind. She had always thought of Emily as a kindred spirit, a fellow struggling artist.

A coldness spread through her limbs. She replayed their conversations in her head, searching for clues, for hints. She found nothing. Just the carefully constructed image of a life lived on a shoestring budget, a life that apparently never existed.

Emotion: bewildered

Cluster: Surprise / Confusion
PC1 (Valence): -1.87 Negative
PC2 (Disposition): -0.48

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.300
ness0.255
像是0.241
worse0.231
Worse0.219
Suppressed:
la-0.373
மகி-0.229
happy-0.226
ecstatic-0.226
K-0.219