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 found the little cafe first. That was always the way. He'd catch a fleeting glimpse of her, a flash of red hair, a quick movement in the crowd, and his world would tilt. He knew the art piece was hers, she’d mentioned it once, a casual aside, a brushstroke of conversation. Now it was on display, large and vibrant, in the window of a bookstore, bathed in the afternoon sun. He needed to be near it. Near her.

He bought a coffee, black, and sat at a table by the window, ostensibly reading a book. But his attention was elsewhere. He kept glancing at the painting, a vibrant landscape, but really, he was looking for her. His pulse pounded a frantic rhythm against his ribs. He felt restless, compelled to move, to speak, to do… something.

She walked past. He recognized the shape of her shoulders, the way she held her head. She paused, and he saw the small intake of breath. He was sure she was going to come inside, and at the thought, a thrilling heat bloomed in his chest. But she didn't. She walked on. A pang of disappointment hit him, sharp and immediate.

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