Michael slammed the desk drawer shut, the force of it rattling the old desk. His grandfather’s things. What had his grandfather been hiding? The letter lay open on the blotter, the script looping and elegant. He read: “… unable to continue…” His vision blurred. He blinked, trying to clear the mist. He reread the words: “…moral objections…”

He paced the room, his jaw clenched. Each step felt thunderous in the otherwise silent study. He couldn’t understand. His grandfather, the stalwart, the dependable, the man who’d built a life around duty… *moral objections*? To what? The clock ticked, each second a hammer blow against his skull.

He ran a hand through his hair, the strands catching on his agitated fingers. He began muttering to himself, a low, unintelligible hum. He picked up the letter again, holding it close to his face. What had they all *missed*? He needed answers, and he needed them now. He could feel it in his teeth, this ache.

Emotion: hysterical

Cluster: Fear / Anxiety
PC1 (Valence): -3.19 Negative
PC2 (Disposition): -0.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.401
0.401
점점0.396
😫0.377
それでも0.373
Suppressed:
la-0.672
a-0.618
de-0.541
😎-0.370
😎-0.359