self-critical - Story 127399 | Emotion Vectors in Gemma4-31B

The sun streamed through the window of the small café, illuminating the dust motes dancing in the air. Maria, seeking a quiet corner, pulled out her newly purchased book. She'd been feeling restless, a vague sense of dissatisfaction gnawing at her. A delicate, folded piece of paper slipped out.

The letter was a shopping list, a grocery list. It was written in a hurried scrawl, but the words were filled with love. The writer mentioned preparing a meal for their children. Maria felt a pang of resentment. Her own life felt empty.

She felt the heat rise in her face as she read the words, the casual domesticity of it all. It felt like a pointed reminder of everything she lacked. She sat there, the weight of her loneliness pressing down on her.

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