The flickering neon sign of the bar across the street blurred, the letters swimming in Elias's vision. He slumped further into the booth, the worn vinyl cool against his cheek. He had been on the road for months, a relentless cycle of gigs, late nights, and instant noodles. A distant piano melody drifted from the stage, a familiar chord progression. He barely registered it until the singer’s voice, a reedy soprano, began the first verse. It was *his* song, the one he’d poured his heart into, the one that had earned him a small but devoted following. A dull ache throbbed behind his eyes. He closed them, rubbing the bridge of his nose.

He’d almost forgotten the feeling of hearing it performed. This wasn't the raw, heartfelt rendition he'd crafted in his own practice room. It was...polished. Bland. The melody felt hollow. Each note felt miles away.

He considered leaving, but the effort of standing, the walk to the door, felt monumental. Instead, he signaled the waiter for another beer. He took a long drag of the new beer as the song continued. He felt a wave of emptiness wash over him.

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