Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

He wandered through the house, each room seeming to echo with a strange, unsettling quiet. He'd been looking for her journals, but instead, he’d found the treasure trove. A box from elementary school, middle school, high school. His entire history, meticulously documented, collected, and guarded.

He leaned against the kitchen counter, his legs feeling unsteady. A wave of confusion washed over him. He felt as if he had lost his footing on solid ground. He picked up a drawing from kindergarten. The memories felt both distant and overwhelmingly present, a cacophony of sensations he couldn't quite place. He was adrift in a sea of his own past.

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