self-critical - Story 127908 | Emotion Vectors in Gemma4-31B

He woke up with a start, heart hammering in his chest. The news had been a nightmare, but the reality was even worse. He grabbed his phone and re-read the email, each sentence a blow to his already fragile confidence. The sunlight, streaming through the blinds, felt harsh and unforgiving.

He dragged himself out of bed and went to the mirror, looking at his face. He felt like he had failed everyone, starting with himself. He should have listened to his parents, should have taken a different path. He was incapable.

He ran the water, splashing it on his face. The coldness did not diminish the burning sensation in his chest. His reflection in the mirror seemed to stare back at him with disgust, confirming his fears. He was the problem.

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