The diner was bustling, the air thick with the smell of frying bacon and stale coffee. She was trying to focus on the crossword, to distract herself. Then, from the jukebox, a guitar intro that was all too familiar. Her song, but distorted, mangled. A karaoke version. The singer's voice grated, the tempo rushed.

She felt a heat rise in her face. The coffee tasted bitter. She wanted to shout, "Stop!" but the words wouldn't come. She focused on her trembling hands, trying to calm them. The jukebox’s lights flashed, a garish mockery of the feeling she'd poured into the lyrics. She dropped a crumpled napkin on the table and ran, escaping the song.

Emotion: grief-stricken

Cluster: Sadness / Despair
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -1.14

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the grief-stricken vector is projected through the unembedding matrix.

Promoted:
😞0.396
泪0.338
😢0.332
haunting0.331
😔0.325
Suppressed:
de-0.434
😎-0.330
B-0.303
🤓-0.285
😎-0.276