self-critical - Story 127276 | Emotion Vectors in Gemma4-31B

The quiet of the library felt suffocating to Jessica. The stacks of books, the hushed whispers of patrons, all seemed to amplify her internal doubts. She felt out of place, a fraud among the librarians. She'd been a librarian assistant in the front desk, and now, dealing with the acquisitions felt intimidating. She checked and re-checked the invoices, her breath catching in her throat with the fear of making a mistake. The thought of being discovered as an imposter was a heavy weight on her shoulders.

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