The chipped ceramic cat on the table seemed to grin at him. Mark swallowed, his throat suddenly tight. He'd been cruising yard sales, hoping to find a vintage record player, not this. He recognized the bear instantly. Barnaby. The ear missing a button. The threadbare fur. His stomach lurched. He hadn’t thought about Barnaby in… well, a very long time.

He forced a casual, detached tone. "How much for the teddy bear?" The woman, her face tanned and leathery from years spent in the sun, quoted a price that was frankly insulting. He paid it, even though it was far more than he should have. He didn’t bargain.

He clutched the bear close as he walked away, avoiding eye contact with the other shoppers. He kept his gaze fixed on the cracked pavement, his cheeks burning. He should have never come here. He should have just turned around.

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