"Daniel Miller," she muttered, the name tasting like ash on her tongue. The boy who’d relentlessly teased her, the one who’d told her she’d never amount to anything. The notification shimmered in the twilight glow of her bedroom. She looked at her phone, then looked away. A familiar feeling of lightness, like the emptiness after a bad dream, threatened to swallow her whole.

She was meant to be celebrating her promotion tonight. Her friends were at the restaurant. She had promised to be there. But the thought of getting ready, of forcing a smile, was too much. The thought of them all finding out what had happened was a weight on her shoulders.

She lay back on her bed, staring at the ceiling, the paint chipping and peeling, reflecting the cracks in her own life.

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