remorseful - Story 116247 | Emotion Vectors in Gemma4-31B

The news conference was a disaster. Janet, a politician, had been "diagnosed" with a terminal illness in what was perceived as a calculated ploy. Now, the revelation of a mix-up left her speechless.

She felt the heat rise in her cheeks, her face burning. The press, usually so supportive, were now baying for blood. She couldn't meet their eyes. She felt the weight of their judgment.

Her campaign manager, Mark, approached her. His usual cheerful demeanor was gone, replaced by a tight-lipped expression. She felt a terrible pang of… something. She had hurt him, and she knew it.

She retired to her office and stared at the framed photos. She felt like she had betrayed everyone.

Emotion: remorseful

Cluster: Shame / Guilt

PC1 (Valence): -1.57 Negative

PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`S`	0.656
`L`	0.580
`悔`	0.463
`remorse`	0.416
`been`	0.398

Suppressed:

`de`	-0.444
`(!)`	-0.309
`имеется`	-0.289
`oldukça`	-0.287
`😍`	-0.285