The stairs seemed impossibly steep. Each step was a battle, her legs heavy and uncooperative. The rejection letter, crumpled in her purse, felt like a lead weight dragging her down. She fumbled with the key, the metallic click a jarring sound in the sudden quiet of her apartment. The scent of stale air and forgotten plans filled her nostrils.

She dropped her bag on the floor with a dull thud. Her reflection in the hallway mirror revealed a stranger – eyes shadowed, hair unkempt, a slump in her posture that spoke volumes. The apartment felt vast and empty. She wanted nothing more than to curl up and vanish, to be swallowed whole by the silence. She slowly began to remove her coat, feeling the mundane movement consume her.

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