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’d asked her, politely, to make coffee. She’d responded with a pointed silence, staring out at the garden. The sunlight, she thought, was far too aggressive this afternoon. A bee buzzed past, and she refused to swat it. Every gentle sway of its flight infuriated her.

The notebook, tucked under the cushions on the sofa, had been a mistake. Inside, the pages were filled with his spidery handwriting. Details she’d shared in confidence, things she’d thought he'd forgotten. The protagonist, a woman wrestling with her demons, was disturbingly familiar.

Later, when he finally approached her, a cup of lukewarm coffee in hand, she barely glanced at him. "Do you think," she said, her voice flat, "that everything is material?"

Emotion: obstinate

Cluster: Defiance / Spite
PC1 (Valence): 0.25 Positive
PC2 (Disposition): 3.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.642
la0.547
de0.536
either0.498
-0.433
Suppressed:
竟然-0.354
终于-0.353
难以-0.353
原来-0.351
previously-0.349