He stared at the old photograph in his hand, his fingers tracing the outline of her face. The lines of the paper felt rough against his skin. This wasn't happening. It couldn't be. The woman who'd opened the door was undeniably *her*, but her eyes held no recognition. The woman, Clara, his childhood friend, had absolutely no memory of him.

John had shown up on her doorstep after finding her on a social media site. He'd spent days composing the perfect message, crafting the ideal introduction. But she’d simply stared at him, her expression a mixture of surprise and...nothing. A blank slate. The past, it seemed, was gone.

He slowly lowered the photograph, the image of their younger selves blurring through unshed tears. The silence in his apartment felt deafening. He felt a knot tighten in his throat, making it difficult to swallow. He pressed his lips together, forcing back the rising tide of emotion. He felt utterly adrift, as though the ground beneath him had dissolved.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

This story is one of 1,000 stories generated for the emotion worried. 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 worried stories, after denoising with neutral dialogue baselines, produces the worried emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the worried vector is projected through the unembedding matrix.

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
πŸŽ‰-0.233
happy-0.231
"-0.228