The crumpled paper felt rough against Leo's trembling fingers. He’d found it tucked away in the back of a drawer – a list, neatly handwritten in Amelia's familiar script. "Swim with dolphins." "See the Northern Lights." He swallowed hard, the lump in his throat making it difficult to breathe. He’d thought he knew her, known their shared dreams, but these felt… different.

He sat on the bed, the duvet twisting around his legs. The room seemed too quiet, the absence of her humming a tangible thing. He wished she was here, even just to argue about the dishes, just to break the suffocating stillness. He ran a hand through his hair, the strands catching on his fingers as if trying to restrain him.

He reread the words, each one a sharp jab to the heart. “Learn to play the ukulele.” Amelia had always loved music, but he’d been too preoccupied, too focused on his own anxieties, to truly notice. He felt a deep ache within his chest, a hollow space that echoed with unsaid words and missed opportunities. He missed her already.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322