Michael stared at the list, the paper crinkling in his sweaty palm. It was tucked inside his father’s old wallet, found amongst the other things he was sorting through. "Visit the Amazon rainforest," it said. Then, underneath that, "Learn to play the guitar, properly." He felt a cold sweat prickle across his scalp. His father, a man of routine and comfort, had written these things.

He ran a hand through his hair, the strands catching on his suddenly rough skin. The next entry was innocuous: "See the Northern Lights". The final one, written in a shaky hand, made him want to bolt from the room: "Go shark diving." He pictured his father's cautious smile, and a wave of nausea rolled over him. He felt as if the room was shrinking, and it was hard to breathe.

He slammed the wallet shut. He couldn’t. He just couldn’t. It was all a mistake.

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309