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 email arrived like a gift. "Important Announcement Regarding [Building Name]," the subject line read. John practically vibrated with anticipation as he clicked it open. His eyes quickly scanned the text and a broad grin stretched across his face.

He let out a low chuckle, a sound of profound delight. He could feel his heart beat faster. He imagined the interior renovations, the value of the property going up. He could finally plant those roses he'd been dreaming of on the balcony.

He immediately reached for his phone, dialing his financial advisor. “Great news,” he practically bellowed into the phone. The news gave him an immediate surge of energy. This was a win.

Emotion: pleased

Cluster: Positive / Joy
PC1 (Valence): 3.62 Positive
PC2 (Disposition): -0.28

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.728
la0.496
happy0.462
🤩0.433
🥳0.430
Suppressed:
S-0.486
😞-0.481
ness-0.454
😣-0.431
worse-0.401