"Is that...?" Chloe trailed off, her voice barely a whisper. She stared at the dog-eared phrasebook on the living room table. It lay open to a section on idioms, the words in a language she hadn’t spoken since she was a child. The language that represented her family, her roots. The language she kept at a distance.

Mark sauntered into the room, beaming. He looked…smug. “Surprise!” he declared, his voice buoyant. He’d clearly been practicing. The words he’d just uttered, in her native tongue, had a forced quality, like he was delivering a line in a play.

She felt a flicker of something close to frustration. He was trying to connect with her heritage, but the way he was doing it felt…wrong. Forced. Inauthentic. She turned away, suddenly consumed by a need for the emptiness of the front porch.

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