self-critical - Story 127995 | Emotion Vectors in Gemma4-31B

The university sent out a celebratory email about its most successful applicants. And there it was, his essay, "The Language of the Mountains," as an example of excellence. He felt a sharp pang of… something. He wasn't sure what it was. But it made him feel tight in the chest.

He immediately went to the essay and reread it. All the things he liked about it then, the elegant phrasing, the vivid descriptions, now seemed forced and artificial. It felt like a costume he had once worn, now discarded and exposed.

He went out for a run. His legs pumped, his breath ragged. He ran harder, pushing himself until the fatigue numbed the feeling in his chest.

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