The phone vibrated insistently on the table, its shrill buzz a physical assault on her ears. Olivia ignored it. She stared at the open laptop screen, at the scathing blog post she was drafting. Her fingers hovered over the keyboard. She felt a knot in her stomach, a tight, unpleasant sensation.

The allegations against Seraphina Bellweather, the author she'd idolized, were true. Olivia had spent weeks analyzing the evidence, the sources, comparing the novels. She knew the truth, and she had to tell it.

She hadn't slept well in days. Every time she closed her eyes, she saw the stolen phrases, the unacknowledged inspiration. Each tap of the keyboard was a deliberate act of defiance. Yet, each keystroke was a painful hammer blow against the pedestal where she had placed Seraphina. She closed her eyes. Then she hit "publish."

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