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 been so secretive, lately. Always on the phone, always rushing off somewhere. She'd dismissed it as work. Then, she saw the bank statement. A joint account. One that was her name only, with a balance she didn't recognise. Her stomach lurched.

She scrolled through the transactions, each one a slap in the face. She wanted to scream, to lash out, to break something. She felt a burning sensation in her chest, a suffocating pressure.

She thought of their vows, their promises, their shared future. They were all meaningless now. She felt a deep, cutting loneliness, as though she were adrift in a vast, cold ocean, utterly alone.

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