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 will reading was short, the atmosphere tense. Elias and his sister, Maya, had always been rivals. Their inheritance reflected this. Elias received the family's thriving tech company, a legacy built on innovation and hard work. Maya, however, received a collection of antique cameras and a darkroom tucked away in the basement.

She went down to the darkroom, the air stale and heavy. Looking at the equipment, the chemicals, she felt a profound sense of… exclusion. It was a language she didn’t speak, a world she didn't belong to.

Elias called her later, his voice booming with congratulations. "I'm thinking of upgrading the servers. We should do a joint venture." She found herself clenching her jaw. She replied, "I don't think so."

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274