tense - Story 147127 | Emotion Vectors in Gemma4-31B

The man's brow furrowed. He was trying to read, but the words swam before his eyes. He slammed the book shut, his teeth grinding. His son's championship game. He was supposed to be there. He felt the weight of the delay pressing on him, crushing the hope inside his chest.

He glared at the departure board, each updated time of departure like a punch to the gut. He ran a hand over his face. He felt like his blood was boiling. The airport was a prison, the flight a mirage. He stood up, pacing back and forth, his gait jerky and uncoordinated.

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