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).

Rain lashed against the windows of the hospital waiting room. Arthur sat, slumped in a vinyl chair, watching the water cascade down the glass. The persistent rhythm lulled him into a state of serene detachment. He felt no urge to fidget, no anxiety about the news they were about to receive.

A woman with fiery red hair and a kind face approached. She hesitated, then sat down next to him. Her movements were slow, deliberate. "They told me to wait," she said, her voice a low murmur. "It’s… overwhelming, isn’t it?"

Arthur simply nodded, unable to summon more than a gentle inclination of his head. He recognized the feeling, this pervasive sense of calm, this quiet acceptance of whatever fate held in store. He knew what she meant; the news, whatever it was, seemed unimportant.

Emotion: docile

Cluster: Passivity
PC1 (Valence): 0.25 Positive
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ค่อย0.311
淡淡0.302
enjoy0.273
enjoyed0.273
สบาย0.272
Suppressed:
la-0.589
de-0.413
l-0.373
!!-0.344
!!!-0.316