The silence in the apartment was deafening, punctuated only by the rhythmic ticking of the grandfather clock in the hallway. It was unbearable. Sarah slumped onto the sofa, the plush cushions offering little comfort. The phone, which had been buzzing with calls for the past few weeks, sat innocently on the coffee table. She had started to notice his behaviour, how he would disappear to take calls in the other room.

She wanted to scream but all that came out was a quiet whimper. She felt a profound sense of abandonment. He would come home later, and everything would be fine, she told herself. But the words felt hollow and meaningless.

The clock ticked on. Time moved forward without her.

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