“Another rejection?” Mark’s wife, Sarah, asked, her voice laced with a weary sympathy that grated on his nerves. He hated the pity in her eyes. He had been applying for jobs for months, each rejection letter deepening his sense of… futility. He was perpetually on edge. He could feel the pressure in his temples, a constant throb.

He slammed the door to his home office, the sound echoing through the otherwise quiet house. Inside, the walls felt like they were closing in. He stared at the newly installed hardwood floor, the source of a recent, very large debt.

He started digging through boxes, searching for a misplaced tax document. He found a dusty file labeled “Historical Records.” Curiosity piqued, he opened it, and his world shifted. A faded newspaper clipping described a grisly murder that had occurred in the house decades earlier. His stomach churned. The weight of it all was almost too much.

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