The professor's voice crackled through the classroom speakers. He’d been expecting a lecture on the French Revolution. Instead, the professor’s face was grim. “The university will be closing at the end of the semester.” A collective gasp rippled through the room. He felt a sinking feeling, not just of disappointment, but of something deeper, heavier.

He watched the faces of his classmates. Some looked shocked, some upset. He looked down at his own hands, the nails bitten to the quick. He hadn't been taking his studies seriously enough. He hadn't worked hard enough. He hadn't networked, hadn't sought out opportunities. He had wasted his time.

He felt the familiar pressure in his chest, the internal tightening that always accompanied his failures. He thought back to the late nights spent playing video games, the classes he’d skipped, the projects he'd half-heartedly completed. His regret was a bitter taste in his mouth.

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:
L0.748
S0.662
ness0.425
ly0.374
0.351
Suppressed:
la-0.492
soon-0.304
(!)-0.301
secured-0.284
cautiously-0.280