self-critical - Story 127707 | Emotion Vectors in Gemma4-31B

The overhead lights buzzed, mimicking the relentless anxiety that had been plaguing Jessica all day. She replayed the meeting in her head, cringing at her contributions. She was sure she had said the wrong things, that she had sounded foolish.

Across the room, Michael was laughing, animatedly describing something. He always seemed so confident, so at ease. Jessica felt a fresh wave of unease.

“Whispering Pines, huh?” Michael said, turning to face her. “Small world.”

Jessica felt her cheeks flush. She hadn't thought about that camp in ages.

"Yeah," she mumbled. She'd struggled at Whispering Pines, constantly comparing herself to others.

"I loved the campfires!" Michael said, reminiscing. "Remember the ghost stories?" He was reminiscing, sharing a happy memory. She just remembered the dark nights. She suddenly found the corner of her desk incredibly interesting.

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