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 old, yellowed paper crackled in her hands. She couldn't stop grinning. Her mother’s bucket list. A simple thing, yet it had unlocked a vault of forgotten dreams. Her chest felt light, like a balloon about to float away.

"Learn to dance." The idea was so unexpected, so… fun! She imagined her mother, graceful and elegant, twirling across a ballroom floor. It made her ache for a little bit of her mother to come back.

She immediately started researching dance classes. A quick search later, she had already signed up. The prospect of learning to dance ignited a spark within her. She was giddy with anticipation, a smile permanently etched on her face.

Emotion: thrilled

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): 0.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.613
!0.544
B0.508
🤩0.499
🎉0.457
Suppressed:
S-0.601
😞-0.536
ness-0.462
😥-0.351
😔-0.341