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).

A sickness rose in David’s throat. He'd poured his heart, his secrets, his very soul into those sculptures. Now, they were on display in the town square, massive and imposing. They’d been stolen from his studio months ago, but he’d hoped, prayed, that they wouldn't be found.

He hid behind a towering oak, watching the townsfolk marvel at his creations, unaware of their true origins. His stomach churned. He wanted to scream, to shout the truth, but his voice seemed trapped in his chest. A knot formed in his throat, making it difficult to breathe.

He gripped the rough bark of the tree, his knuckles white. The light of the setting sun cast long, distorted shadows, mirroring the grotesque figures in the square. A shudder racked his body. He felt a desperate urge to run, to disappear into the gathering darkness.

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