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 beach was deserted, the sand cool and soft beneath her feet. Chloe walked along the shoreline, the rhythmic crash of the waves a soothing sound. The vastness of the ocean always made her feel… okay. She inhaled the salty air, feeling the sun warm her face. Her brother, Ben, approached, looking serious.

“Chloe, I have to tell you something." Chloe stopped walking, but she smiled.

"I've become a Wiccan."

Chloe looked at the horizon. She smiled.

"That's great. What did you think of that new movie about Wicca? I thought it was pretty good." She resumed her walk, the waves gently lapping at her toes. “Want to go get some ice cream later?”

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350