Michael froze, his hand halfway to a dusty, tarnished teapot. His breath hitched in his throat. Across the jumbled collection of forgotten treasures sat it. The bear. He knew its every imperfection, the way its left paw was slightly smaller than the right, the patch on its belly where he’d tried to repair a tear with clumsy stitches. It was Theodore, his constant companion throughout a tumultuous childhood.

His pulse hammered against his ribs. He felt a sudden, inexplicable heat in his cheeks. The woman running the sale was chattering to a potential buyer, oblivious to the emotional maelstrom brewing.

He took a step closer, feeling as though he was walking through molasses. He knew the woman selling the bear. She was the woman who had lived across the street from him when he was a child. He tried to speak. He tried to form a question. Nothing came out.

Emotion: shocked

Cluster: Surprise / Confusion
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.298
😰0.268
sickening0.266
не0.266
que0.257
Suppressed:
enthusi-0.283
开心-0.269
ولة-0.262
ارك-0.257
উৎসাহিত-0.255