The coffee cup trembled in Leo’s hand. He'd just seen a familiar username pop up in the local coffee shop's Wi-Fi list. “CosmicCat77,” the handle of his online confidante for the past year. His heart did a frantic drum solo against his ribs. He pictured the sleek black hair, the mischievous grin he’d only seen through a webcam. He had never shared his location. Now, here it was, staring him in the face. He clutched his phone tighter, knuckles bone-white.

He ducked behind a towering bookshelf, pretending to browse titles he'd never read. Every rustle of a magazine, every footstep, made him jump. A sudden, cold sweat erupted on his forehead. CosmicCat77 was probably just here for coffee, right? Nothing to be concerned about. He closed his eyes, took a shaky breath, then peeked cautiously over the top of the bookshelf.

The bell above the door chimed. He flinched.

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
🎉-0.233
happy-0.231
"-0.228