The fluorescent lights of the clinic hummed, a monotonous drone that seemed to burrow into Amelia's skull. Each blink felt like a monumental effort, her eyelids heavy and resistant. She slumped further into the stiff chair, the vinyl cold against her skin. The doctor’s voice, when he finally called her name, sounded distant, muffled.

“Amelia, please come in.” He gestured to a chair. "I have your results."

She dragged herself to her feet, legs feeling like they were filled with concrete. The doctor’s smile didn’t quite reach his eyes, and the air in the small office felt thick, stagnant. He pointed to a graph covered in sharp lines and unfamiliar medical terms.

“There seems to be a… discrepancy,” he began, his voice dropping. She looked at the report. It was NOT hers.

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