tense - Story 147080 | Emotion Vectors in Gemma4-31B

He leaned against the brick wall of the bookstore, trying to appear nonchalant. But beneath his calm exterior, his internal engine was revving at full speed. He felt a constant need to pace, to flee. The smell of old paper and ink usually soothed him. Today, it felt oppressive.

He'd spent countless nights in online chats, sharing his life with a woman he thought he knew. When she stepped from the entrance, his heart hammered against his ribs. He felt the blood drain from his face, and felt oddly sick.

He opened his mouth to speak, but the words wouldn't come.

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