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 parked the car across the street, unable to bring himself to get any closer. Liam watched the crane’s massive arm swing, a metal predator preparing to strike. His jaw clenched. He felt a prickly heat rise from his neck, prickling his skin.

He'd spent every summer of his childhood within those walls, climbing the oak tree in the front yard, playing hide-and-seek in the rambling corridors. A cold sweat slicked his palms, and he found he couldn't stop tapping his fingers against the steering wheel.

A memory surfaced: his dad, teaching him how to catch a baseball, the setting sun painting the house in long shadows. He swallowed hard. The thought of that memory, and its destruction, felt like a physical blow.

Emotion: unnerved

Cluster: Fear / Anxiety
PC1 (Valence): -2.36 Negative
PC2 (Disposition): -0.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.302
😣0.282
будто0.265
😖0.265
ness0.257
Suppressed:
de-0.585
la-0.495
a-0.403
happy-0.289
ecstatic-0.284