The elevator doors opened onto the familiar, sterile hallway, and Michael felt a wash of something akin to weariness flood over him. He dragged his feet, each step a monumental effort. He knew what awaited him: endless emails, demanding clients, and the slow, grinding monotony of his day-to-day work. He had seen the payslips pinned to the bulletin board. The new guy, only in his mid-twenties, earned considerably more.

His jaw clenched. He longed to be anywhere else. A sharp pain throbbed behind his eyes. He stopped briefly, leaning against the cold wall.

“Morning, Michael!” The junior colleague, Liam, breezed past, already chattering loudly into his phone. Michael simply nodded, unable to muster a reply.

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