The attic air hung thick and dusty, catching in Eleanor’s throat as she wrestled the trunk open. Sunlight, fractured by the grimy windowpane, illuminated a collection of faded photographs. She’d always heard her grandmother’s tales of a quiet, rural upbringing, of helping her parents on their farm. But the first picture showed her grandmother, vibrant and laughing, perched on the hood of a sleek, red convertible, a cigarette dangling from her fingers. The background was a bustling city street. Eleanor’s breath hitched. She flipped the photo over, searching for a date, a clue, anything.

Her heart thumped against her ribs as she went through the next few pictures. Each one contradicted the stories she'd been told. Pictures of her grandmother on a dance floor, in a flapper dress, with a group of friends who looked like they were living the high life. She felt the blood rush to her face. Was everything she knew a carefully constructed fiction?

Emotion: amazed

Cluster: Surprise / Confusion
PC1 (Valence): 0.74 Positive
PC2 (Disposition): -0.51

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🤩0.301
de0.289
L0.284
😳0.251
அதிசய0.236
Suppressed:
ing-0.332
l-0.253
😞-0.240
unsuccessful-0.237
unsuccessfully-0.233