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 small, cramped office felt oppressive, the walls closing in on Emily. Across the desk, John, her friend, sat frozen, listening to the professor's accusations. He was accused of copying a section of his philosophy thesis. She hated the situation.

Emily knew how diligent he was, how passionately he researched. The professor, a pompous man, had a reputation for finding faults. He spoke in a cold, detached tone, leaving John to sink further into despondency.

Emily fought back a rising tide of fury. She wanted to scream, to defend him. Instead, she took a deep breath, and quietly asked the professor a question about his methods. It was a small act of rebellion, a small act of support.

Emotion: compassionate

Cluster: Compassion / Love
PC1 (Valence): 3.19 Positive
PC2 (Disposition): 1.62

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.811
B0.618
H0.550
sympat0.422
latter0.422
Suppressed:
此刻-0.519
не-0.514
솟-0.446
désormais-0.446
仿佛-0.444