The office seemed to shrink around David. His mentor, Ms. Evans, always said the office was too large, impersonal. Now, it swallowed him whole. The empty desk, stripped of all personal items, offered no solace. The only reminder of her was an expired sticky note, a gentle admonishment, still clinging to the monitor.

He avoided the water cooler, the social hub where she'd always had time for a chat. He found himself avoiding his colleagues. He felt a deep, pervasive heat rise up in his neck. He worked through his lunch hour, lost in the monotonous task of data entry.

He replayed their last conversation. He cringed at his own arrogance. He'd been too busy, too focused on his own advancement. He felt a sudden, sharp pang in his chest.

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