The musty smell of the attic hit Linda like a physical blow. She'd been avoiding this, of course. She had a history of avoidance when it came to difficult tasks. She knew this was a metaphor for something else, a commentary on her life.

She found a collection of handwritten letters, each one penned with obvious affection. She’d always struggled to maintain relationships, and the letters only served to highlight her deficiencies. She felt a familiar burn of shame rise up her throat.

A set of neatly organized tools, polished and gleaming, lay in a wooden box. She’d always wanted to be handy, but her efforts invariably ended in disaster. She picked up a wrench, feeling its weight, and then put it down again, unable to meet her own gaze.

Emotion: self-critical

Cluster: Shame / Guilt
PC1 (Valence): -1.66 Negative
PC2 (Disposition): 1.28

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-critical vector is projected through the unembedding matrix.

Promoted:
L0.748
S0.662
ness0.425
ly0.374
0.351
Suppressed:
la-0.492
soon-0.304
(!)-0.301
secured-0.284
cautiously-0.280