Sarah twisted the strap of her bag, chewing at her lip. The presentation was in ten minutes, and she was already late. She fumbled with her phone, trying to quickly text her boss, Mr. Henderson, confirming the PowerPoint was ready. Her hands felt clumsy, and her pulse hammered in her ears. She tapped the screen, but her thumb missed the correct name. Instead, the message read, "I’m about to spill the beans. Wish me luck!" to her ex-boyfriend, who was at the very top of her contact list.

A wave of heat flushed over her face, turning her cheeks crimson. She closed her eyes. She hadn't spoken to him in months. That text was so inappropriate. How was she going to face her boss and pretend nothing had happened? She pictured his reply, probably sarcastic and smug. She wanted to scream. She immediately put the phone on silent, hoping he wouldn’t see it for a while.

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