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).

Liam was whistling as he walked into the bakery, the scent of fresh bread already making his stomach rumble. He’d just finished a successful gallery showing, and a sense of lightness propelled him. He practically bounced on the balls of his feet as he ordered a croissant and coffee.

His phone buzzed. It was a DM from an unknown account. He clicked on it, curiosity piqued. The image showed a shoulder blade, and on it, a nearly perfect replica of his charcoal sketch, “Solitude.”

A whoop escaped him. He felt like he could float. He imagined the person, carrying a piece of his work with them, etched onto their skin. The feeling was electric. He paid for his treat with a skip in his step, already planning his next project.

Emotion: joyful

Cluster: Positive / Joy
PC1 (Valence): 3.96 Positive
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.615
de0.600
!0.572
excitedly0.552
enthusi0.533
Suppressed:
S-0.893
ness-0.513
😞-0.470
😣-0.337
L-0.334