tense - Story 147138 | Emotion Vectors in Gemma4-31B

The professor’s words, sharp and accusatory, sliced through the air. “Your work bears a striking resemblance…” Mark felt a flush creep up his neck. He shifted in his seat, a nervous twitch in his eye.

The room seemed to spin slightly. His heart hammered a frantic rhythm against his ribs. He crossed his arms, hugging himself, as if seeking some measure of protection.

He stared at the ground, unable to meet the professor’s gaze. The silence stretched on, a suffocating blanket that wrapped itself around him. He felt utterly and completely exposed.

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