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 world seemed to open up before Chloe as she watched her son, Noah, run towards the school building. The air felt charged with potential, and a smile played on her lips. She felt a lightness within.

"Who do you have for a teacher, Noah?" she called out, her voice filled with a happy anticipation. Noah, turning back, yelled, "Mr. Miller!" Chloe's breath hitched. Mr. Miller?

A wave of warmth, like a familiar embrace, wrapped around her. Mr. Miller, her own favorite teacher, a man who had helped shape her into the person she was. A feeling of pure, unadulterated happiness bloomed in her chest.

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