The humming of the elevator, usually a background noise, now became an irritant, a constant reminder of her helplessness. She felt herself becoming increasingly agitated, her jaw clenched tight. The presentation was in an hour, and she was nowhere.

She tried to take a deep breath, but her lungs seemed to have forgotten how. Her heart was racing. She needed to leave.

She stared at the floor numbers, willing the indicator to move. She ran her hands through her hair, smoothing it back, then pulling at the ends. She kept repeating herself, silently. "It will be okay. It will be okay."

She finally just leaned against the wall, her eyes squeezed shut, trying to shut out the world.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🤩-0.210