The aroma of stale beer hung heavy in the air of the empty bar. Sarah leaned against the bar, the cool wood a welcome contrast to the internal heat of her body. She’d been running on fumes for weeks, and her body ached. The waitress, a bored woman with tired eyes, seemed to be moving in slow motion.

"That road trip… best time of our lives, right?" Mark asked, twirling the straw in his empty glass. His voice had a lighthearted quality that felt foreign to her. "Remember the Grand Canyon? The sunrises? How we had to get towed by a farmer because our car broke down?"

Sarah closed her eyes, trying to recall the shared memories. "We went to the beach. Remember? We barely had any money for anything; we camped in that tiny tent, and it rained the whole time." She found it difficult to speak, as if a great weariness had taken over her.

Mark's brow furrowed. “The beach? No, we drove for days… across the desert. We were stuck for hours, talking to that friendly old farmer, who fixed our car!” He said, his voice rising with disbelief.

Sarah felt a hollow ache. The beach, the salty air, the damp clothes… it was all imprinted in her memory. "I… I don't know," she whispered, the words barely audible, her head hung low.

Emotion: worn out

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.12 Negative
PC2 (Disposition): -1.20

Role in Research

This story is one of 1,000 stories generated for the emotion worn out. 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 worn out stories, after denoising with neutral dialogue baselines, produces the worn out emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the worn out vector is projected through the unembedding matrix.

Promoted:
😞0.265
😩0.234
疲れ0.226
que0.221
sighed0.221
Suppressed:
B-0.357
because-0.228
l-0.222
భి-0.208
あくまで-0.208