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 click and whir of the camera, the thrill of capturing the perfect moment – these were the things that defined David’s life. He had spent his time as a photographer. Now, his son, seven-year-old Noah, was the one behind the lens.

David found Noah in the backyard, crouched in the grass, his eyes glued to the viewfinder. Noah was taking photos of a ladybug. He was completely absorbed, his tongue poking out from the corner of his mouth. David watched, a strange, delightful tension building within him. The shutter clicked rapidly. He felt the urge to grab his own camera, to participate, to be a part of this world again.

“I got it, Dad! I got the shot!” Noah called out.

Emotion: stimulated

Cluster: Alertness
PC1 (Valence): 0.38 Positive
PC2 (Disposition): -0.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.317
T0.312
intensity0.280
S0.271
inesis0.265
Suppressed:
a-0.640
l-0.434
la-0.392
que-0.378
fact-0.354