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 flight is delayed," the announcement crackled. Michael's initial disappointment was short-lived. A thrill coursed through him. He was a photographer, and the wedding he was meant to document was now in limbo. He felt a sudden surge of inspiration.

He grabbed his camera and began wandering the airport, his eye searching for beauty. The mundane world transformed. He found a unique angle for his photographs.

He captured moments of joy, of sorrow, of quiet reflection. He felt a newfound freedom, a vibrant connection to his craft. The missed wedding was a catalyst, a chance to see the world with new eyes. This was exactly what he needed.

Emotion: invigorated

Cluster: Positive / Joy
PC1 (Valence): 3.28 Positive
PC2 (Disposition): 0.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.452
(!)0.439
🎉0.434
invigor0.416
enthusi0.395
Suppressed:
😞-0.524
😔-0.428
😢-0.404
😥-0.393
que-0.391