The fluorescent lights of the office seemed to hum louder than usual, a high-pitched whine that grated on Sarah’s nerves. She’d overheard a snippet of conversation, just a stray phrase, from the new hire, young Kevin, bragging about his starting salary. It was significantly higher than her own, after five years of dedicated service.

She kept replaying the scene in her mind, the confident way Kevin tossed his head, the way he laughed. Sarah found herself staring at her keyboard, the familiar keys seeming foreign and mocking. Her fingers hovered over the email draft she was working on, suddenly unsure how to even string a sentence together.

A coldness spread through her chest. Was she missing something? Had she somehow failed? She felt the urge to call her husband, to talk it through, but the thought of forming the words, of expressing the strange disquiet churning inside her, felt utterly impossible.

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