"More tea, dear?" Her aunt’s voice, saccharine and cloying, grated on Chloe’s nerves. She forced a smile and shook her head, focusing on the stack of photos spread across the table. Another forced smile. Another family gathering.

The photos were from her great-grandmother’s youth. The stories described a woman of severe rectitude, a pillar of the community. In the pictures, however, she was dancing in a speakeasy. She was flirting with men in a way that defied the family's tales.

Chloe felt a rising heat in her cheeks, a flush of anger and something else, something she couldn’t quite name. She excused herself, the lie a practiced reflex, and went to the bathroom to get away. The mirrored image of her own face felt unfamiliar. She ran cold water over her wrists, trying to quell the inner storm.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199