The attic air was thick with dust and the scent of forgotten things. Sarah dragged her feet across the creaky floorboards, each step an effort. A long afternoon of sorting through her grandmother’s belongings had her feeling leaden. Her eyelids felt like they were weighted, and the corners of her mouth sagged downwards. She stumbled upon a box tucked away in a dark corner, its contents promising escape from the tedium.

Inside, nestled amongst yellowed lace and brittle postcards, were photographs. One caught her eye: her grandfather, beaming and young, draped over the hood of a sleek, black car. A car her mother had always insisted he’d never owned. The photograph felt like a secret, a gentle prod to unravel the inconsistencies. Her grandmother had always regaled her with the story of their humble beginnings, a tale of hard work and thrift. This photo… this was something else.

Emotion: sleepy

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.86 Negative
PC2 (Disposition): -1.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😴0.318
weary0.310
fatigue0.296
💤0.283
似乎0.280
Suppressed:
B-0.343
C-0.294
because-0.293
భి-0.281
เพราะ-0.277