“Another one,” Mark muttered, his voice a low rasp. He’d been fielding these for a week. Childhood bullies, all of them, crawling out of the woodwork on social media. Each one a fresh reminder of the years he'd spent feeling small and insignificant. This one was Brenda. He recognized the name instantly. He barely had the strength to check, but his fingers did it anyway.

He felt leaden, every joint protesting the simple movement of swiping through her profile. The photos showed a happy life, a life free of the daily dread he’d endured. A life he hadn’t had. He felt hollowed out, a shell of himself. He considered ignoring it, deleting the request, but his hand hovered over the accept button. He wanted to know, to see, but his body just wouldn't co-operate. He just stared.

Emotion: worn out

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.12 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.265
😩0.234
疲れ0.226
que0.221
sighed0.221
Suppressed:
B-0.357
because-0.228
l-0.222
భి-0.208
あくまで-0.208