The morning sun, usually a welcome companion, felt like a harsh spotlight on Amelia's face. She dragged herself out of bed, each movement a monumental effort. Her limbs felt heavy, and a dull ache throbbed behind her eyes. The coffee machine gurgled, a sound she typically found comforting, but today it just added to the general din of the world. Then, she saw it: "The Unexpected Life of Amelia Blackwood," gleaming mock-leather on the coffee table.

A jolt, like an electric shock, shot through her. Where had this come from? Flipping it open, she saw her name, her photograph, and… details. Intimate details about her childhood, her relationships, her secret fear of heights. Her breakfast, the toast she had planned to enjoy, sat abandoned on the counter.

Later, she found herself staring into the abyss of her reflection in the bathroom mirror. The woman looking back was a stranger, yet intimately known. How could someone she didn't know possess such a complete understanding, or at least a carefully constructed perception, of her existence? The weight of it all was almost too much.

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