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 bustling marketplace overwhelmed Clara. The cacophony of vendors hawking their wares, the press of bodies, the overwhelming aroma of spices – it was all too much. She clutched her head, a throbbing headache blooming behind her eyes. It was a feeling she knew all too well. It was the precursor to a bad day.

She saw him standing near a stall, his face pale, his movements halting. He looked just as frazzled. He made eye contact, and his expression was a mirror of her own.

He pushed through the throng toward her, and said "I need to sit down." He then pointed to a nearby bench, and added, "Do you want to?"

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228