self-critical - Story 127736 | Emotion Vectors in Gemma4-31B

He stared at the email on his phone, the words "Austin, Texas" blurring before his eyes. It was from Emily, announcing her move. He felt a familiar knot tighten in his chest. He was the problem.

He scrolled through their old texts, picking apart every word, analyzing every emoji. He always seemed to say the wrong thing, to not be there enough. He wasn't surprised that he was getting left behind.

He slammed his phone on the table. The silence of the apartment felt deafening. He felt a surge of frustration. He should have been more, done more, said more.

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