He found himself avoiding eye contact with the construction workers. They were friendly enough, waving and smiling as they began to dig the post holes. He'd even offered them ice water, a gesture he immediately regretted. He mumbled something about the heat before turning and scurrying back inside.

The sight of the newly erected fence sections, stark and imposing, brought a knot to his stomach. His lawn, which he’d let get away from him, looked even worse in comparison. The uneven patches of crabgrass, the fallen leaves he hadn't bothered to rake, seemed to scream their neglect.

He considered going to apologize for the mess. But each time he started to walk across the lawn he turned back. He knew what he’d find—Mrs. Peterson, his neighbor, glancing over, her expression unreadable. He could almost hear the unspoken judgement. He sat on his porch, and he watched the birds, unable to meet the eyes of the workers.

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