tense - Story 147187 | Emotion Vectors in Gemma4-31B

The constant buzz of the washing machine seemed to amplify the silence in the house. Michael sat on the couch, staring blankly at the television screen. He had overheard the tail end of a call this morning. He wished he hadn't.

He watched as his wife, Jessica, walked in. She was humming, a song he didn't recognize. He felt the blood drain from his face. She looked almost… carefree.

"What's that song you're humming?" he asked, trying to keep his voice neutral. Her smile faltered for a fraction of a second, and he knew.

Emotion: tense

Cluster: Fear / Anxiety

PC1 (Valence): -1.32 Negative

PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`S`	0.213
`😰`	0.191
`느껴`	0.183
`😣`	0.183
`😖`	0.181

Suppressed:

`a`	-0.260
`de`	-0.258
`la`	-0.256
`happy`	-0.218
`B`	-0.198