Sarah dragged herself out of bed, the sheets clinging to her legs. Another Monday. Her apartment felt heavy, the air thick with the remnants of last night's takeout. She longed to stay under the covers, lost in the quiet darkness. But then she remembered Liam, her online friend. They’d been chatting for months, sharing everything from silly memes to serious life anxieties. He was always there, a comforting presence.

Later, scrolling through her phone during her commute, she saw Liam had posted a picture of a coffee shop. "Best latte in the city!" the caption read. A flicker of recognition sparked in her. The exposed brick, the quirky artwork… it was the coffee shop down the street from her office. She messaged him, a playful, “Fancy seeing *you* there!”

His reply was immediate: “No way! You’re local?” The subsequent messages flew back and forth, culminating in a tentative suggestion: “Coffee this week? My treat.” A small smile finally touched her lips, a rare moment of lightness in the perpetual gloom.

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