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 old photographs spilled across the table, like scattered secrets. Amelia, breathing quickly, felt a strange lightness as she examined them. The family stories painted her grandfather as a reserved, intellectual man, a pillar of the community.

But the photographs told a different tale. One picture showed him dressed as a clown at a fair, the face of a man completely different from the family's picture of him. His eyes shone with a mischievous glint that she'd never witnessed. Another showed him laughing with a group of friends, a woman nestled close to his side. The official stories had always been so boring.

She felt a giddy excitement, a deep craving to understand what this man was like, to piece together the fragments of his hidden life. She wanted to know if that woman was still alive.

Emotion: infatuated

Cluster: Compassion / Love
PC1 (Valence): 1.61 Positive
PC2 (Disposition): 0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.525
😍0.421
🤩0.418
😍0.410
उसे0.397
Suppressed:
que-0.427
怨-0.330
a-0.311
도-0.307
愤-0.299