The lawyer’s office was opulent, the air thick with the scent of leather and ambition. He sat across from the man, the author, who was explaining the research process. “I assure you, Mr. Harding, everything is accurate.” The author’s voice was smooth, reassuring. Harding’s jaw clenched. The man’s words were like a foreign language, devoid of meaning. He rubbed the bridge of his nose, the headache growing. He felt his face flush as his eyes darted around the room. The lawyer’s questions, polite and probing, felt like needles. He gripped the edge of the chair, knuckles white.

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