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 doorbell rang. It was David, early, with the cake. Emily stared at him, her heart sinking. The party was off. She hadn't been able to call him yet. "Come in," she managed, her voice tight. "We need to talk."

David, sensing the shift in the air, entered cautiously, the frosted cake swaying precariously in his grasp. Emily relayed the news, the words falling from her lips in a hushed torrent, describing the venue's sudden closure. She felt her face flush hot, then cool.

David’s expression softened. He didn’t interrupt or offer any trite platitudes. He simply placed the cake on the kitchen counter, then turned to her, his gaze kind. "Okay," he said softly. "So, what now?" The weight of their disappointment settled between them, a tangible presence. Emily paused. They looked at each other. She had another idea.

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