self-critical - Story 127154 | Emotion Vectors in Gemma4-31B

He paced the living room, the worn rug a familiar comfort. He'd rehearsed his answers a hundred times, and still, he felt as though he'd stumbled over his words. Liam, a born orator, would have nailed it.

He had promised himself he wouldn't compare himself to Liam. Yet, here he was, replaying the interview, cringing at every perceived misstep. He picked at a loose thread on the sofa.

The phone rang. He hesitated. It was probably Liam, calling to gloat, or maybe to offer hollow words of comfort. He took a deep breath, and answered.

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