tired - Story 151107 | Emotion Vectors in Gemma4-31B

The constant tapping of keyboards, the ringing phones, all echoed in Clara's skull. The words on the email, *Training of New Hire: Clara to assist with the transition*, seemed to vibrate with a particular intensity. She tried to focus on the task at hand – training Maya – but it was like wading through molasses.

Maya, young and full of life, was eager to learn. She asked insightful questions, which made Clara feel the pressure.

Clara just nodded along, struggling to keep her eyes open. She tried to force a smile, but it felt wooden and unnatural. It was hard to keep the tone up. She really needed a break.

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