tense - Story 147180 | Emotion Vectors in Gemma4-31B

The beach felt wrong. The waves crashed with a relentless energy, mirroring the turmoil in Kevin's chest. He stared out at the ocean, unable to focus. He had been so proud of his best friend, Daniel, the supposed "film director."

Now, he was at Daniel's family home, and the "film set" was a shed in the back garden, strewn with cheap equipment.

Daniel joined him, squinting in the sunlight. "Beautiful day, huh?" he said, his voice strained. He shoved his hands in his pockets, his posture slouched.

Kevin took a deep breath. The ocean felt vast and unforgiving.

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