self-critical - Story 127316 | Emotion Vectors in Gemma4-31B

"Did you finish that report?" Anya's voice was flat, as she walked in. She found Mark, her partner, on the balcony. He was talking on the phone, and as he spoke, she felt a wash of anger as he uttered words she understood. A word she had just learned.

The phone conversation ended abruptly, and Mark turned to face her. "Hey," he began, and the look in his eyes was instantly regretful. Anya felt her cheeks heat up. She wanted to scream.

She moved toward the kitchen, grabbing a glass and filling it with water. The ice cubes clinked and cracked.

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