A wave of fatigue washed over Sarah as she sat slumped in her chair, the air conditioning doing little to combat the oppressive atmosphere. The digital clock on her monitor glared back at her; 4:57 PM. Another three minutes until she could leave and face the traffic. Her shoulders ached with a persistent tension she couldn’t shake.

She'd been meticulously planning a new presentation, pouring hours into it, when a casual comment from Mark, her junior, shifted everything. "Yeah, the new bonus structure is great. Really helps the bottom line.” The implications struck her like a physical blow. She realized the junior colleague made far more than her despite her years of service.

The world seemed to tilt on its axis. Suddenly, the emails felt heavier, the meetings interminable. Even the simple act of typing seemed to require a herculean effort. She stared at the cursor, blinking, wanting to crawl into a dark room and sleep for a week.

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