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).

“I think I’ve got it,” Sarah’s daughter, Chloe, announced, her fingers dancing across the keys of the old piano. Sarah, a composer herself, had always dreamed of passing on her passion.

Sarah watched Chloe, her small hands moving with a grace that surprised her. Chloe’s rendition of a complex Chopin piece flowed, the notes precise, the interpretation insightful. She’d only been playing for five years, but was already mastering the classics.

Sarah felt a surge of… something. A mixture of pure joy and a fierce protective feeling. She’d laid the foundation, but Chloe had built a cathedral. She had surpassed her, already. She was going to be better than Sarah had ever been.

Emotion: valiant

Cluster: Defiance / Spite
PC1 (Valence): 2.82 Positive
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.887
willing0.543
la0.452
/0.444
👊0.439
Suppressed:
S-0.661
마다-0.518
😞-0.488
😔-0.461
😩-0.444