The coffee cup trembled slightly in her hand as she tried to explain the nuances of the client database. Her stomach felt like a knotted ball of wire. “So, when you input the data… you have to be *very* careful with the formatting,” she stammered, aware that her voice was higher than usual.

Across the worn conference table, the woman, Deborah, sat poised and ready. Her posture radiated an unflappable confidence that aggravated her. She was younger, clearly ambitious, and already seemed to know more than she let on.

She glanced at the window, desperate for the meeting to end. The sunlight seemed unusually bright today. It felt as if it was spotlighting all of her flaws. She hoped this was the last time she would ever see this room again.

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