self-critical - Story 127528 | Emotion Vectors in Gemma4-31B

The newspaper lay open on the kitchen table, the headline screaming: "Literary Giant Exposed." Across the room, Jason watched his wife, Emily, stare in silence at the article on her phone. The article detailed how her favorite writer, Marcus Thorne, had apparently stolen the plot of a short story she had written in college.

He’d always known about Emily’s writing, how much it meant to her. She hadn't written anything in years. The house felt empty without the gentle tap, tap, tap of her keyboard. He wanted to reach out, to comfort her, but the words felt trapped.

He reached for his cup of coffee, but his hand shook so badly he spilled it.

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