self-critical - Story 127724 | Emotion Vectors in Gemma4-31B

The scent of cedar hung heavy in the cedar chest in the spare room. Liam knelt, running his fingers across the smooth wood, a soothing gesture he didn't realize he did. He’d promised his sister he’d sort through their father’s belongings.

He pulled out a file, and his blood ran cold. A draft resignation letter. The letter was dated the year his parents had almost divorced. He knew now why he always felt he didn't measure up.

His jaw clenched. He was always disappointing. He was sure he caused his parents’ issues. He felt the familiar heat creeping up his neck. He wanted to shout, to break something.

He shut his eyes, took a shaky breath, and, instead, began to meticulously organize the papers in the file, seeking order in the chaos, a futile attempt to control the emotions churning within.

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