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 scent of freshly baked bread wafted from my open kitchen window, a comforting aroma on a Saturday morning. I poured myself a cup of coffee and strolled onto the back porch. Mrs. Rodriguez, always dressed in vibrant colors, was already waiting at the property line, a basket of ripe tomatoes in her hands.

"Good morning, Mr. Evans!" she greeted. She explained she needed a bit more space for her tomatoes. Her garden always overflowed with life.

I took a slow sip of coffee, savoring the warmth that spread through me. Her enthusiasm was infectious. I considered the proposal carefully, then looked at the vacant space.

"Certainly," I said, with a nod. "Why don't we get those tomatoes planted right away?"

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