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 attic air hung thick with dust motes dancing in the lone sunbeam. Clara had expected tears, the crushing weight of grief. Instead, a lightness bloomed in her chest as she sifted through the old trunk. Her mother’s handwriting, neat and familiar, scrawled across the paper: “Climb a mountain. Learn to salsa.” She laughed, a genuine sound that surprised even herself. The list wasn't a lament; it was a map.

The thought of salsa lessons made her smile. Her mother, always a terrible dancer. The trunk clanked shut. The oppressive gloom of the attic didn't feel so heavy anymore. She could almost *see* her mother grinning.

Emotion: relieved

Cluster: Positive / Joy
PC1 (Valence): 2.22 Positive
PC2 (Disposition): -0.97

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.638
finally0.606
finalmente0.597
relieved0.561
enfin0.546
Suppressed:
l-0.443
:(-0.439
😣-0.429
S-0.388
😖-0.385