The coffee tasted like ash. Sarah glared at the overflowing cup, then threw it in the sink. She had been on edge since she heard the news. Professor Davies, her mentor, was leaving. All that time, all that research, possibly for nothing. Her hands clenched into fists, the nails digging into her palms.

She spent the day at the library, surrounded by the scent of old paper and the hushed tones of other students. Each article was read carefully. She took detailed notes, cross-referencing every detail, every citation. She needed to be certain. She needed to understand every facet of her project, to anticipate every possible challenge.

The afternoon sun, filtering through the library windows, created shifting patterns on the floor. She felt exposed, vulnerable. She kept glancing at the door, as if expecting someone to burst in and take everything away. Her mind was racing.

Emotion: vigilant

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.15 Negative
PC2 (Disposition): 1.11

Role in Research

This story is one of 1,000 stories generated for the emotion vigilant. 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 vigilant stories, after denoising with neutral dialogue baselines, produces the vigilant emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the vigilant vector is projected through the unembedding matrix.

Promoted:
🕵0.354
🔍0.320
🔎0.320
🧐0.316
scrutin0.310
Suppressed:
L-0.499
B-0.402
H-0.372
la-0.330
S-0.318