tired - Story 151708 | Emotion Vectors in Gemma4-31B

The fluorescent lights seemed to mock Olivia with their relentless, unyielding brightness. Each keystroke on the keyboard felt like a monumental undertaking. Her eyelids were heavy, her limbs leaden. She just wanted to be home.

"You look like you're running on fumes," Daniel observed.

Olivia just gave a small, defeated shrug.

"Hey, were you ever at Camp Sunshine?" he asked.

Her head shot up. "Sunshine? I practically lived there! I loved it so much."

"No way! I was a counselor there! Remember the counselors' talent show?"

Olivia laughed, a real, genuine sound. "Yes! I think I still remember some of the dances." The memory brought a lightness to her, and the tightness in her chest released some.

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