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 scent of oil paint and turpentine hung heavy in the air, a familiar comfort. Eleanor watched her daughter, Maya, meticulously layering crimson onto a canvas. The same canvas that used to be hers. A wave of warmth washed over her. It had started with Maya asking to watch her paint, then to help, and now, at thirteen, the girl possessed a talent that surpassed Eleanor's own. Seeing Maya's focused brow, the way her small hand danced with the brush, stirred a feeling deep within her chest. It was a feeling of genuine pleasure to see her passion.

Eleanor set about making a cup of tea, humming a quiet tune. She added two sugars, just the way Maya liked it.

Later, Eleanor found Maya, still working, lost in her own world. A new painting had already begun and was taking shape.
"Want a snack, sweetie?" Eleanor offered gently. Maya nodded, her eyes still fixed on the canvas, and then smiled a genuine smile.

Emotion: thankful

Cluster: Positive / Joy
PC1 (Valence): 3.50 Positive
PC2 (Disposition): -0.83

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
happy0.493
மகி0.487
grateful0.461
mutlu0.452
feliz0.431
Suppressed:
[]-0.482
😖-0.478
😣-0.467
😫-0.461
worse-0.450