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 morning sun splashed across Amelia’s face, warming her like a happy hug. She practically bounced to the door when she heard the insistent rapping. Mr. Henderson, his face crinkled with concern, pointed a shaky finger at the towering oak that straddled their property line.

"It’s getting a bit… precarious, Amelia," he croaked, his voice barely audible above the birdsong. "I was wondering, would you mind if we took it down?"

Amelia threw her head back and laughed, a bright, clear sound that startled a flock of pigeons into flight. “Down? Absolutely!” she exclaimed, clapping her hands. “I’ve been thinking of a new patio anyway! We can have a grand unveiling, a block party even! The possibilities!”

She was already imagining fairy lights strung through the branches of a smaller, more manageable tree. The air seemed to shimmer, charged with the sweet scent of possibility. She couldn't stop grinning.

Emotion: exuberant

Cluster: Positive / Joy
PC1 (Valence): 3.42 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.676
de0.605
la0.596
!"0.551
!0.548
Suppressed:
S-0.892
😞-0.474
ness-0.407
L-0.389
😔-0.376