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 box of old photographs lay on the table, its contents a jumbled mess of memories and unfamiliar faces. Sarah, sifting through the faded images, found one tucked away in the back – a picture of her mother with a man she didn't recognize, her hand resting on his arm. It was a photo from before Sarah was born.

A wave of unease washed over her. She dropped the picture, her hands suddenly clammy. The walls of her small apartment seemed to press in on her, and she felt a sudden need to flee. She pushed the box away from her, the sound of the cardboard scraping on the table a harsh interruption of the silence.

Emotion: afraid

Cluster: Fear / Anxiety
PC1 (Valence): -2.76 Negative
PC2 (Disposition): -0.70

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.316
마다0.314
worse0.313
😰0.303
恐惧0.290
Suppressed:
de-0.691
la-0.562
a-0.402
"-0.316
brag-0.293