self-critical - Story 127198 | Emotion Vectors in Gemma4-31B

The back-to-school night was approaching. *Mrs. Chen, First Grade.* John dreaded the thought. Mrs. Chen. The teacher who had constantly compared him to his older brother. The teacher who had told him he was not smart. His ears burned just thinking of it.

He spent the evening obsessing over his son, Alex’s, reading skills. He drilled him relentlessly. He pushed him to memorize every word. Alex, who had previously loved reading, started avoiding books.

He was suddenly obsessed with making sure everything was correct. He would make sure that Alex was studying. He did not want Alex to make any mistakes.

He remembered how his confidence had been shattered. He hated it. He felt the weight of his own inadequacies. He wanted Alex to be perfect.

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