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 vibrations in the wall were undeniable. The new home owners, the Millers, were gutting their kitchen. Maria was already used to the noise. She made herself a cup of chamomile tea, its calming aroma filling the air.

She picked up her knitting. The click of the needles provided a soothing counterpoint to the distant sounds of demolition. Each stitch was a small victory, a step toward completion. She was making a blanket for her grandson.

She thought of the Millers, of their new kitchen, and their future. She took a deep breath, focusing on her work. She would take it one row at a time. The blanket was coming along nicely.

Emotion: patient

Cluster: Passivity
PC1 (Valence): 2.71 Positive
PC2 (Disposition): 0.40

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
rest0.467
eventually0.415
leisurely0.369
慢慢0.357
a0.355
Suppressed:
无法-0.432
even-0.428
인해-0.421
無法-0.415
😣-0.391