The office felt cold, despite the thermostat. Liam hunched over his computer, trying to ignore the persistent tremor in his hands. He'd just sent a draft of a report to his supervisor, and the familiar wave of worry had crashed over him. He was sure it was riddled with errors. He could already see his boss's disapproving expression.

"Camp Whispering Pines," Sarah said from across the aisle. "That was the best summer ever!"

Liam froze. He suddenly felt exposed. Whispering Pines was a place of painful memories.

"You were there too?" he asked, his voice flat. He had always struggled to compete. He remembered the humiliation of losing the swimming competition.

"Totally! I loved all the activities. Remember the talent show?" she asked excitedly. He remembered his terrible magic act. Sarah was now sharing how she had won a race. He felt a familiar sense of deflation. He looked at his hands, finding them suddenly difficult to control.

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