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 string quartet’s rendition of “Here Comes the Bride” grated on David’s nerves. He hadn’t been expecting to see her, not tonight. Seeing Chloe, standing near the bar in a shimmering dress, her eyes sparkling, felt like a physical blow. A tightness formed in his chest.

He stalked over to the bar, ignoring the well-wishers and the joyous chatter. When the bartender asked what he wanted, he barked, “Scotch. Neat.” He took a long, burning gulp, the alcohol doing little to soothe the internal storm. He knew he should leave, but something kept him rooted to the spot, watching her.

Emotion: hostile

Cluster: Anger / Hostility
PC1 (Valence): -1.67 Negative
PC2 (Disposition): 1.56

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.592
S0.387
骂0.324
est0.301
aggravated0.295
Suppressed:
a-0.457
de-0.375
H-0.289
🤩-0.280
la-0.273