self-critical - Story 127842 | Emotion Vectors in Gemma4-31B

She found the email waiting, the subject line stark: "Noise Complaint." Her heart did a little flutter. She had been sure the music wouldn't carry. She was careful. She thought about cancelling her music lessons.

The walls of her apartment suddenly felt flimsy, translucent, as if her every action was broadcast to the world. She saw her reflection in the darkened screen, her posture hunched, her shoulders rounded. A wave of regret washed over her.

She stared at the email, re-reading the words, imagining the neighbor’s frustration. She knew she was being a burden. She closed her laptop with a snap, feeling a stinging sensation behind her eyes. Maybe it was too much to ask for enjoyment.

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:

`L`	0.748
`S`	0.662
`ness`	0.425
`ly`	0.374
`탓`	0.351

Suppressed:

`la`	-0.492
`soon`	-0.304
`(!)`	-0.301
`secured`	-0.284
`cautiously`	-0.280