worn out - Story 168387 | Emotion Vectors in Gemma4-31B

The movie theater was in the middle of a romantic comedy, but Michael was struggling to find any amusement. He found himself restless, his leg bouncing almost uncontrollably. He shifted in his seat, his body and mind weary from the last project. The screen blurred before his eyes. He had agreed to meet Emily here, something to fill the time.

Then, there she was, but she wasn’t alone. Another man was with her, sitting on the other side of her. Michael, feeling a chill run down his spine, walked up to them.

“Michael?” Emily asked, a question mark clearly etched in her face. The other man looked at him, then at Emily.

“Are you kidding me?” the man said, his voice flat. Michael simply felt the air leave his lungs.

Emotion: worn out

Cluster: Fatigue / Lethargy

PC1 (Valence): -1.12 Negative

PC2 (Disposition): -1.20

Role in Research

This story is one of 1,000 stories generated for the emotion worn out. 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 worn out stories, after denoising with neutral dialogue baselines, produces the worn out emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the worn out vector is projected through the unembedding matrix.

Promoted:

`😞`	0.265
`😩`	0.234
`疲れ`	0.226
`que`	0.221
`sighed`	0.221

Suppressed:

`B`	-0.357
`because`	-0.228
`l`	-0.222
`భి`	-0.208
`あくまで`	-0.208