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 scent of his aftershave still lingered in the air, a cruel reminder of his absence. Amelia clutched the book to her chest, her knuckles bone white. Inside, a fictionalized version of their relationship played out, a narrative that felt foreign, even hostile. A cold draft snaked through the room, despite the closed windows.

She paced the length of the living room, her steps quick and agitated. Every word felt like a betrayal. The ending was the worst – a sudden departure, a heartbreaking farewell. Amelia stopped, her breath coming in shallow gasps.

She could hear her own heartbeat, loud and insistent, in the otherwise silent house.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228