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).

The musty smell of the antique shop clung to him as he held the faded photograph. His grandfather, a man known for his quiet dignity, was pictured in a boxing ring. There was a look of brutal determination on his face. The family had always said he was a scholar, a teacher.

He felt the blood drain from his face. His heart began to race. The image challenged everything he thought he knew about his grandfather. The photograph was from a time before he was born, and it suggested a secret life.

The shop owner's voice, asking if he was alright, seemed to come from a long distance. He felt as though he were walking through a dream. He just wanted to go home and lie down. He felt a sudden, profound exhaustion.

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