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 email from Mark, the neighbor, sat in her inbox, a digital request for space to cultivate vegetables. Amelia read it again, each word a provocation. The idea of sharing her backyard, her hard-earned sanctuary, ignited a fire within her.

Her fingers tightened around her coffee mug. She imagined the chaos, the weeds, the inevitable encroachment. She thought of the pristine green of her lawn, a canvas she'd perfected over years of dedication. A knot formed in her stomach.

She tapped her foot impatiently. She closed her eyes for a moment, picturing the vegetables that would be growing, and the vegetables she would be missing out on. "Absolutely not," she muttered under her breath, a smile of quiet defiance creeping across her face.

Emotion: stubborn

Cluster: Defiance / Spite
PC1 (Valence): 0.80 Positive
PC2 (Disposition): 3.73

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.780
l0.685
la0.605
either0.580
a0.501
Suppressed:
难以-0.435
终于-0.419
顿时-0.405
竟然-0.396
wondering-0.391