tense - Story 147139 | Emotion Vectors in Gemma4-31B

It was the waiting that was the worst. Ava’s leg bounced erratically, a frantic drumbeat against the floor of the waiting room. Each tick of the clock was a hammer blow against her already fragile composure.

She had rehearsed her defense a hundred times, but the words felt hollow, inadequate. Her mouth felt like a desert, and she kept swallowing, trying to moisten her parched throat.

She took a deep breath, attempting to steady herself, but the air felt thin and difficult to grasp. The door to the professor’s office opened, and her stomach lurched. It was time.

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