Rain lashed against the windows of the old Victorian. Clara slumped further into the threadbare armchair, the silence of the house pressing in on her. The air smelled of dust and forgotten things, a scent that clung to her throat and tightened her chest. This was all she had now: this crumbling house, and the memory of her grandmother, a woman whose smile seemed to mock her from beyond the grave.

Her brother, David, was whistling a jaunty tune in the kitchen, probably admiring the peeling wallpaper. He always saw the potential; she saw the wreckage. Clara felt a knot form in her stomach.

She dragged herself up, her limbs heavy. The walk to the kitchen felt like wading through treacle. β€œWhat are you so chirpy about?” she grumbled, her voice flat. David just grinned, oblivious.

Emotion: miserable

Cluster: Sadness / Despair
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.87

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.369
无法0.321
焑法0.270
πŸ’”0.265
πŸ˜”0.262
Suppressed:
de-0.490
la-0.329
B-0.300
l-0.292
/-0.224