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 streamed through the open window, warming Elara’s face. She stretched, a long, luxurious movement that pulled every muscle in her body taut before releasing. A small smile played on her lips as she skipped into the kitchen, the scent of freshly brewed coffee already filling the air. Her mind was buzzing with possibilities, ideas bouncing like energetic pinballs. She'd been up all night, sketching and tinkering, lost in the pure joy of creation. Now, ready for a new day, she felt the kind of electric energy that made her want to dance. Then, the email arrived: "Patent Filed - [Elara's Invention Name]" by a competing company. A wave of ice coursed through her, stealing the joy from her body.

Emotion: refreshed

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.998
de0.703
!).0.384
(!)0.381
!".0.379
Suppressed:
S-0.782
C-0.478
L-0.414
😣-0.388
😞-0.376