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 silence of the backyard was deafening. I felt a chill despite the warm afternoon, a deep, pervasive cold that settled in my bones. My eyes scanned the area; there, where the apple tree had been, was a circle of dirt and splintered wood. Its branches, where I used to swing as a child, were just memories. I hadn't been back in years.

I backed away involuntarily, my feet moving as if guided by an invisible force. The air felt thick, heavy, as if the very atmosphere had been altered. My hands trembled, my stomach churned. The world seemed to shrink, the edges of everything blurring.

The emptiness screamed.

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