Sarah slumped onto the park bench, the late afternoon sun doing little to warm her chilled bones. It had been a long week, punctuated by sleepless nights and a persistent knot of anxiety in her stomach. The notification had pinged just as she was trying, and failing, to enjoy her chamomile tea. "Your teenage diary has been published online." The words, stark and unforgiving, still swam before her eyes. She ran a hand over her face, the rough stubble of her morning shadow a surprising comfort against her palm. Each sentence, each carefully crafted adolescent angst-ridden observation, now dissected and judged by the world.

She looked around the park, at the families laughing, the dogs chasing frisbees, the elderly gentleman feeding pigeons. A profound sense of isolation washed over her. It wasn't just the invasion of privacy, it was the raw exposure of her most vulnerable self. She clutched her phone, a tiny, cold weight in her hand, a portal to a past she now desperately wanted to disown. The silence of her apartment, which she had once craved, felt now like an echoing prison.

Emotion: weary

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.52 Negative
PC2 (Disposition): -1.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.384
😞0.288
sighed0.230
ness0.210
😪0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
🤟-0.198