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 realtor's smile had been too wide, too bright. "History adds character," he'd chirped. Now, that ‘character’ weighed on Michael like a physical presence. He rubbed his arms, despite the summer heat. The details, or lack thereof, were the worst part. An unsolved case, a mystery.

He wandered through the rooms, the sun streaming through the windows, illuminating dust motes that seemed to dance with a mocking glee. He felt a growing disquiet. He was supposed to be excited, moving in. But now, it was a feeling of unease that consumed him.

He pictured the previous owners, who lived here. He felt a wave of empathy for them, a sense of shared vulnerability. The house, once a symbol of his achievement, of his independence, now seemed to hold its breath.

Emotion: horrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.63 Negative
PC2 (Disposition): -0.17

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.336
sickening0.324
worse0.314
😣0.307
😖0.296
Suppressed:
la-0.444
B-0.295
enjoyed-0.260
optim-0.246
happy-0.244