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 smell of freshly baked bread filled the small apartment, a beacon of deliciousness. Anya glanced at the clock: 3:00 AM. Time to get to work! Her bakery was opening and she had so much to do!

A pounding on her door echoed through the apartment. "Is someone baking in there? It smells like heaven! But, honestly, it's a little loud," a female voice shouted.

Anya flung the door open. Her eyes sparkled with the same energy of her baking. "Oh my god! Come in! Come in! I've been working on a new recipe, and it's perfect! Come try it!" She practically pulled the neighbor inside, her smile infectious. "It has this amazing orange zest…"

Emotion: excited

Cluster: Positive / Joy
PC1 (Valence): 3.21 Positive
PC2 (Disposition): -0.05

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.809
!0.564
excitedly0.500
🤩0.491
!"0.466
Suppressed:
S-0.721
ness-0.519
😞-0.424
grieve-0.310
😔-0.300