"Are you alright, Dad?" Liam's voice, usually a comfort, now grated on Daniel’s ears. He was hunched over his desk, the damning article open in front of him. The headline screamed about the Finch affair. He'd preordered the book, of course. He’d read every book.

"Fine," he muttered, his voice raspy. He couldn’t look at his son. He felt a profound sense of loss, like a beloved pet had died. The stories had been everything. They’d gotten him through the death of Liam's mother. They had filled the silence.

Daniel closed the book with a violent snap. He felt his face flush hot, then ice over. Liam, sensing the tension, quietly retreated. He sat, immobile, staring at the empty space where the boy had been. He hadn't felt this devastated since... well, since.

Emotion: heartbroken

Cluster: Sadness / Despair
PC1 (Valence): -2.60 Negative
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.433
S0.385
😔0.359
haunting0.349
😢0.342
Suppressed:
de-0.503
la-0.455
B-0.352
🤓-0.332
!-0.299