tense - Story 147837 | Emotion Vectors in Gemma4-31B

A chill ran down Robert’s spine as he reviewed the printout of the bank statement. He tried to think, but his thoughts scattered. He felt a sudden, inexplicable nausea. He leaned against the kitchen counter, trying to regain his equilibrium. His wife, Emily, had always been so careful, so pragmatic with their finances. This didn’t fit.

He picked up the phone, wanting to call his friend. He hung up. He knew what he *should* do, but he felt unable to make the first move. The phone seemed heavy in his hand. Time seemed to slow.

The front door opened. He jumped.

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