The news landed like a lead weight in Leo’s stomach. Across the country. Across the country. The words echoed in his skull as Maya spoke, her voice bright, full of excitement, while Leo felt his face grow hot. He managed a strained smile, a pathetic imitation of the happiness he knew he should be feeling.

He picked at a loose thread on the couch, avoiding eye contact. He knew he should say something, anything, to validate her feelings, but the words caught in his throat. This wasn't a good move for him.

Later, walking home, the streetlights seemed to mock him, casting long, distorted shadows that mirrored the twisted feeling within. He found himself kicking a discarded soda can, the metallic clang a small, pathetic expression of his inner turmoil.

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
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🀩-0.242