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 typed list from his grandfather sat on the kitchen table. “Build a birdhouse” was the very first item. Daniel, a city dweller, had always considered himself unskilled. Yet, as he stared at the list, he experienced a strong surge of excitement, a feeling that had been missing from his life.

He bought wood, a hammer, and nails. His fingers got sore, and he got frustrated as he worked. But with each nail hammered and each piece of wood joined, a sense of satisfaction grew.

When the birdhouse was finally complete, imperfect but solid, he felt a powerful sense of accomplishment. It was an awkward, beautiful thing that reminded him of a man who’d tackled challenges with his own hands.

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