Sarah tapped her pen against the conference table, a staccato rhythm that grated on her boss's nerves. The patent rejection notice was tucked inside her briefcase, a constant, unwelcome presence. She stared at the presentation slides, the carefully crafted images mocking her.

Her stomach churned. The team had lauded her invention, believing it to be groundbreaking. Now, everything felt like a lie. She caught her breath, then forced a smile as her boss asked a question about her device.

She swallowed. "The competitor's claims are... ambiguous, sir." Her voice sounded strained, brittle. She kept her fingers fidgeting, unable to stay still. The meeting seemed to drag on forever.

Later, she sat in her car, hands gripping the steering wheel so tight her knuckles were white. The city lights blurred through her tear-filled eyes. How could she compete?

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:
S0.213
😰0.191
느껴0.183
😣0.183
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198