"Unbelievable," Clara muttered, the word a small, lost sound in the vast silence of her study. She'd just finished reading the report: her beloved author, Julian Thorne, had lifted entire scenes from her unpublished novel, "Whispers of the Coast." She couldn't breathe.

She reread the side-by-side comparison, her eyes tracing the lines, each word a betrayal. Her stomach twisted; a cold sweat broke out across her forehead. The man who had shaped her worldview, whose books she had reread countless times, was nothing more than a fraud. She was almost unable to move.

She felt a desperate need to scream, to lash out. The anger, the shock, the profound sense of loss, threatened to overwhelm her. She slammed her fist on the desk. The sharp crack echoed in the room, and she gasped. The room felt smaller now, the walls closing in, suffocating her.

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