Sarah slumped against the bus stop shelter, the chipped paint cold against her cheek. Rain lashed against the glass, blurring the already indistinct streetlights. She’d been staring at the same page of her textbook for an hour, the words swimming before her eyes. The application essay. *That thing*. Now, here it was, in the university's official guide, the very words she’d agonized over, held up as some paragon of admission essay excellence. A hollow chuckle escaped her. Wasn't that just her luck?

"Excuse me," a voice broke through her reverie. A woman with an umbrella was looking at her, a polite smile on her face. "Are you alright? You look a bit…well, you look like you need to sit down." Sarah just nodded, her jaw tight. The woman moved over, and Sarah watched the woman’s face as the woman sat. It was the face of someone who understood.

Emotion: tired

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.92 Negative
PC2 (Disposition): -1.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
weary0.254
fatigue0.249
😴0.238
sighed0.237
barely0.230
Suppressed:
C-0.268
because-0.225
B-0.219
🤟-0.217
భి-0.214