ashamed - Story 9525 | Emotion Vectors in Gemma4-31B

“I… I don’t know what to say,” Emily stammered, her voice cracking. Across the table, her mother was beaming. “Silas Blackwood just emailed me!” she’d announced, her face lit with pride. She held up her phone, displaying a reply to an email she'd sent the author about how much Emily loved his work.

Emily stared at her plate, unable to meet her mother’s gaze. The news had broken yesterday. She'd known about the plagiarism, and the irony was a bitter taste in her mouth. She felt a prickling at the corners of her eyes.

She pushed her food around her plate, the familiar flavors suddenly repulsive. She wanted to run, to escape the suffocating weight of this charade.

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