The elevator ride down felt interminably long. Emily gripped the handrail with a fierce desperation, her knuckles bone-white. The numbers flickered past, each floor a ticking reminder of her impending doom. The announcement had been made at the team meeting, the words – “both Emily and Daniel are…” – echoing in her ears.

She felt a sudden, sharp pain in her chest, as if a physical weight had settled there. She couldn't breathe properly. The metallic tang of fear filled her mouth. The elevator doors finally slid open, revealing the brightly lit lobby. She wanted to run, to disappear into the anonymity of the city, to escape the scrutiny of her colleagues.

Her legs felt shaky as she stepped out, as if she were walking on air. The smiles and greetings from the receptionist and security guard seemed mocking, taunting. She kept her head down and hurried towards the exit, her vision blurring with unshed tears.

She needed to get away, to be alone, to think. To *not* think.

Emotion: horrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.63 Negative
PC2 (Disposition): -0.17

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.336
sickening0.324
worse0.314
😣0.307
😖0.296
Suppressed:
la-0.444
B-0.295
enjoyed-0.260
optim-0.246
happy-0.244