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 air invigorated Thomas as he left his house. He felt a sense of… exhilaration. He had been feeling stuck for months, but this morning, the world felt open. He reached into the mailbox and pulled out a plain, white envelope. Inside, a card. It was an anonymous birthday greeting.

He decided to go for a run, something he hadn’t done in ages. The rhythm of his feet on the pavement, the wind in his hair, the sweat on his brow – it was all cathartic. After, he began to make plans to visit his college friends, an idea that had been lost in the fog of his day-to-day routine.

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