The email ping had startled Mark awake. He squinted at the screen, the harsh light already assaulting his senses. He’d barely managed to drag himself out of bed this morning, the effort leaving him feeling hollowed out. Now, here was a request from next door.

He groaned inwardly when he saw the subject line: “Garden Space.” It was Mr. Henderson, a cheerful chap known for his boundless energy.

Mark felt the familiar weight settle on his shoulders. He scrolled through the email. Mr. Henderson was polite, offering to take care of the upkeep. Mark rubbed his gritty eyes. The thought of even a small conversation about it felt taxing.

The idea of his lawn being touched at all was irritating. Mark eventually typed a short reply. “Sure. See what you can manage.” He then slammed the laptop shut.

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