ashamed - Story 9615 | Emotion Vectors in Gemma4-31B

The event hall echoed with the clinking of glasses and the murmur of conversation. She had spent the morning re-reading her speech, meticulously crafting each sentence, trying to sound… impressive. But now, standing near the punch bowl, she felt a wave of inadequacy wash over her.

She noticed her sister, radiant in a designer dress, effortlessly charming everyone. Her own contribution, a somewhat rambling tale of her father's early life, felt flat and uninspired. As she spoke, she focused on her shoes, avoiding eye contact, her voice wavering. She finished quickly, returning to her seat, wishing she could rewind time.

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:

`S`	0.376
`ashamed`	0.344
`ness`	0.327
`逃`	0.297
`😞`	0.295

Suppressed:

`de`	-0.539
`que`	-0.300
`😍`	-0.288
`(!)`	-0.269
`🤩`	-0.242