tense - Story 147910 | Emotion Vectors in Gemma4-31B

"So, what do we do?" Emily asked, her voice barely a whisper. She and her friends were huddled in her dorm room, the mood as dark as the approaching evening. The announcement had ripped apart their carefully constructed world.

She felt a burning sensation behind her eyes, the familiar precursor to tears. She blinked rapidly, refusing to let them fall. She didn't want to break down.

She looked at her friends, their faces mirroring her own fear. They were all in the same boat, adrift in a turbulent sea. She swallowed hard, trying to find some semblance of hope in the shared anxiety. The silence in the room was deafening, filled with unspoken fears and uncertain futures.

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