self-critical - Story 127335 | Emotion Vectors in Gemma4-31B

The aroma of freshly baked bread usually greeted Daniel as he walked into “Le Petit Boulangerie.” Today, the scent was just a hollow memory. He swallowed hard, reading the chalkboard menu that now featured the announcement of its closure. He felt a wave of nausea.

He should have been a better customer. He should have introduced his girlfriend. He was so caught up in the details, so concerned with his presentation to her. Always worrying, always over-analyzing.

He ran a hand through his hair, the gesture conveying his utter frustration with himself. He should have told the baker how much he loved the croissants, told him what a comfort it had always been to him, the small, quiet place. But he hadn't.

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