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 balloons hung limp in the corner of the room, their festive colors somehow mocking the silence. I stared at the remnants of what should have been a vibrant celebration. A birthday party for my best friend, Jessica, was called off due to a sudden change of plans. I felt a familiar pang of hurt, a shadow settling over the cheerful atmosphere.

I imagined Jessica’s reactions. I know how much this meant to her. I knew the value she put in the relationships of our friend group. I knew that she'd be crushed. The ache was familiar.

I grabbed my phone, already composing a text message, my thumbs fumbling over the keys. She needed to know she wasn't alone.

Emotion: empathetic

Cluster: Compassion / Love
PC1 (Valence): 1.55 Positive
PC2 (Disposition): 1.05

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
H0.622
B0.603
L0.529
la0.414
sympat0.393
Suppressed:
iostream-0.441
chẳng-0.393
仿佛-0.359
не-0.352
দুপুর-0.344