He slumped onto the dusty stool, the musty air of the basement clinging to his clothes. Every muscle in his body felt like it was made of lead. The task, sorting through boxes of his mother’s belongings, seemed insurmountable. He rubbed his eyes, trying to clear the fog that clouded his vision. This was a brutal ordeal, and the thought of tackling it all was exhausting.

He reached for a cardboard box overflowing with papers. The top item was a crudely drawn picture of a snowman, dated from when he was six years old. A wave of something akin to sluggishness washed over him. He found himself mesmerized by the simple lines, the bright crayon colors, feeling utterly adrift.

Then, he saw it: a painstakingly crafted report on the rainforest. He barely remembered writing it, but there it was, his handwriting, alongside his mother’s meticulous corrections and encouraging notes in the margins. A heavy feeling settled in his chest, as if a weight was pressing on him. He felt oddly comforted, but also oddly...drained.

He sat there, staring at the paper, the world around him blurring.

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