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 chipped porcelain doll, its painted eyes staring blankly ahead, sat on the mantelpiece. It was the only thing his grandmother had left him, a reminder of her. His brother, Michael, had called, his voice heavy. "Dad needs to sell the doll, Liam. For the mortgage."

Liam felt a knot tighten in his stomach. He pictured the doll in his childhood home. He took a slow breath. He’d known this day might come.

“I’m coming over,” Liam said, his voice even. “Let's see what needs to be done.” He felt a strange sort of peace settle over him.

He was sure that there were options.

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