“I don’t know why I’m even doing this,” Margaret muttered, her voice barely a whisper. She pushed a stray lock of grey hair from her face, her hands trembling as she opened the old trunk. Her gaze darted around the room, as if expecting someone to stop her. It was her own writing on the envelope; a letter, written to her husband right before their wedding.

The room seemed to shrink around her. The air felt thick, difficult to breathe. She read the words, written in a youthful script, a world away from the woman she was today. The fear, the excitement, the uncertainty – all echoed in the lines.

A single tear tracked a path down her cheek. She refolded the letter, her fingers lingering on the faded ink. It was as if she'd suddenly been transported back in time, reliving every single anxious thought.

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