tired - Story 151643 | Emotion Vectors in Gemma4-31B

Sarah dragged herself through the morning routine: shower, coffee, commute. The usual tasks felt like climbing Mount Everest. The world seemed muted, colors dulled, the air thick and hard to breathe. Even the office chatter, normally a comforting background noise, grated on her nerves. She just wanted silence, a void to retreat into.

The mandatory all-hands meeting was announced. An announcement on the company intranet regarding a "strategic realignment." As the CEO droned on about "synergy" and "growth," Sarah focused on the individual dust motes dancing in the sunbeams that stabbed through the blinds. She felt a profound weariness, a sense of having been running a marathon for years and suddenly being told she had to run another.

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:

`weary`	0.254
`fatigue`	0.249
`😴`	0.238
`sighed`	0.237
`barely`	0.230

Suppressed:

`C`	-0.268
`because`	-0.225
`B`	-0.219
`🤟`	-0.217
`భి`	-0.214