He’d walked into the financial aid office with a swagger, confident and sure of himself. Now, standing across from the stony-faced administrator, Leo felt the blood drain from his face. His carefully constructed facade of assurance crumbled, leaving him exposed. He swallowed hard, the lump in his throat making it difficult to speak. He ran a hand through his hair, the gesture more frantic than casual. Outside, the midday sun beat down on the campus, but he felt an icy chill gripping him. The administrator’s voice, a monotone drone, repeated the same damning information: wrong name, wrong student. Wrong. Wrong. Wrong. He wanted to shout, to argue, but the words caught in his constricted chest. He could only manage a weak, "But… how?"

Emotion: shaken

Cluster: Fear / Anxiety
PC1 (Valence): -2.72 Negative
PC2 (Disposition): -0.86

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.351
unable0.269
focus0.266
😞0.258
다시0.257
Suppressed:
de-0.769
la-0.579
a-0.375
happy-0.292
"-0.290