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).

"Condo conversion?" Maria stammered, staring at the landlord, her voice cracking. The man’s practiced smile seemed to widen as he explained the supposed benefits. She could barely focus, the blood rushing in her ears, drowning out his words.

She clutched her purse, her knuckles white. Her apartment wasn't just a place to live; it was her life. This was the only home she’d known in the city. The thought of moving, of finding a new place, filled her with a profound, sinking despair.

She wanted to scream, to lash out. Instead, she managed a weak, "I... I need to think about this," before stumbling out of the office, the clean air outside suddenly seeming thin and difficult to breathe.

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