The dusty, slightly musty scent of the bear, his bear, hit him first. Then the sight. It was there, nestled amongst a pile of chipped teacups and faded children’s books. He didn't think he could feel any worse. He remembered the lies he'd told, the elaborate stories. He'd said he'd outgrown it. He'd said it was lost. He’d acted like the bear was a thing he didn't want, when it had meant the world to him.

"How much for the bear?" he asked, his voice rough. The woman, with her kind eyes and gentle smile, quoted a price.

He pulled out a wad of bills, paying without a word. He didn’t haggle. He couldn’t. He needed to escape. He spun and almost ran towards his car, the bear clutched tightly in his grasp.

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