The coffee tasted like ash in Amelia's mouth. She slumped against the kitchen counter, the ceramic cold against her cheek. Another day, another mountain of emails, another client demanding the impossible. The rhythmic thudding from next door, a new demolition project, was already vibrating through the floorboards. *Just breathe,* she told herself, forcing a deep inhale. The air held the scent of sawdust and fresh-cut lumber, a fragrance that usually filled her with a sense of possibility, but today it just added to the pressure in her chest.

Amelia dragged herself back upstairs, the weight of the day pressing on her shoulders. She looked at the unfinished puzzle on the coffee table. It was supposed to be a relaxing activity, but the pieces felt like tiny, mocking reminders of the projects she'd put off and then left undone. She sighed and started sorting the pieces, hoping the concentration might help.

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
η–²γ‚Œ0.226
que0.221
sighed0.221
Suppressed:
B-0.357
because-0.228
l-0.222
ΰ°­ΰ°Ώ-0.208
あくまで-0.208