He found himself staring at the "Closed" sign on the website. His stomach twisted. The thought of not having another of their meatball subs. He'd put off going to Leo's for weeks, making excuses to himself about dieting and saving money. A knot tightened in his throat.

He slammed his laptop shut, the sound echoing in the empty apartment. He felt a burning discomfort, a need to disappear. He should have gone. He *could* have gone. He knew the owner, Leo, for years. They shared a smile, a friendly nod. Now, it was too late.

He skipped lunch entirely, the familiar gnawing hunger overshadowed by a heavier, more persistent feeling. He couldn't even bring himself to call Leo.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242