The afternoon sun blazed through the window, turning the dust motes dancing in the air into glittering, malevolent specks. Michael stared at the package on the counter. It was marked for "Ms. Vivian Thorne," who had lived here before he had. He scrubbed at the counter with a dishrag, the repetitive motion doing nothing to ease the tightness in his chest.

He paused, the dishrag clenched in his fist. He was never one to snoop, but he felt an odd compulsion to inspect it. He moved slowly. The parcel had arrived a few days ago, but he kept delaying opening it.

He carefully sliced open the tape, revealing a collection of leather-bound journals. Each one filled with elegant, looping handwriting. He found himself immersed in the writing. The details were far more interesting than expected.

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