The elevator doors chimed open, revealing a familiar office lobby. Emily’s breath hitched. She saw him immediately. Daniel, her colleague, was chatting animatedly with the receptionist. He turned, and their eyes met. A wave of ice coursed through her veins.

Her feet felt leaden, rooted to the spot. Her hands trembled, making the papers she clutched rustle noisily. She wanted to turn and flee, to disappear back into the safety of the elevator, but it was too late. He was walking toward her.

He smiled, a wide, friendly gesture, but she couldn’t meet his gaze. She found herself staring at the intricate pattern of the carpet, the vibrant floral design suddenly overwhelming. “Emily, right? Daniel here,” he said, extending his hand. She felt the blood draining from her face.

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