tired - Story 151020 | Emotion Vectors in Gemma4-31B

The dinner table felt vast, the faces around it blurred. The clinking of forks, the murmured conversations, seemed to reverberate through the air. Across the table, her niece, Olivia, announced the news with an almost casual air. Her aunt, Deborah, already felt as though she’d been running a marathon.

Deborah took a slow breath, forcing a smile. The muscles in her face felt tight. She picked at the food on her plate, the flavors muted. The weight of everything – the news, the day, the sheer effort of being present – pressed down on her. The room began to spin. She wanted to be anywhere but there.

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