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 flowers were sitting on the porch when Sarah got home from work. She almost didn't see them, lost in the fog of a long commute. The card wasn't there. Her mind immediately leaped to the worst-case scenario.

She let herself in and locked the door behind her, the click echoing in the suddenly silent house. She slowly approached the flowers, a riot of color that felt somehow menacing. Her hand trembled as she reached for them, a premonition of something terrible settling in her stomach.

She carried the bouquet inside, the weight of it unsettling. The scent, initially pleasant, now felt cloying, suffocating. She couldn't shake the feeling of being watched, even though she was completely alone.

Emotion: rattled

Cluster: Fear / Anxiety
PC1 (Valence): -2.29 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.264
😣0.261
worse0.256
😞0.248
😖0.247
Suppressed:
de-0.571
la-0.414
a-0.377
happy-0.259
enthusi-0.234