The text message – “Big news. Moving to Seattle.” – had landed like a physical blow. Amelia had been staring at the screen for a full five minutes, the words refusing to make sense. She felt the sudden urge to pace, back and forth, the confines of her small apartment closing in. Her breathing came in short, shallow gasps. She had a sudden, overwhelming urge to break something, to scream.

She clutched her phone tighter, the cold metal a small anchor in the swirling chaos. Then, she opened her laptop, the screen reflecting her own stunned expression. She opened her email and then closed it immediately. She opened a document and then shut it without typing. What was she supposed to do now?

The familiar weight of her cat, Mr. Bigglesworth, jumping onto her lap was a small comfort. She stroked his soft fur, trying to find some semblance of calm. She didn't know how she could function.

Emotion: disoriented

Cluster: Surprise / Confusion
PC1 (Valence): -2.65 Negative
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
unfamiliar0.340
ness0.321
失去了0.316
не0.315
будто0.311
Suppressed:
B-0.474
la-0.419
K-0.409
de-0.309
brag-0.302