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, mirroring the storm brewing inside Maya. She paced the length of her tiny apartment, the floorboards creaking a mournful rhythm. A restless energy consumed her. She’d spent the afternoon with Chloe, reliving their senior prom. Chloe had described the night as a carefree celebration, filled with laughter and dancing.

Maya felt her face flush, heat prickling her skin. The memory of her own night was a tangled mess of anxiety and disappointment, hidden shame. She remembered the awkwardness, the humiliation, the crushing realization that she hadn't belonged. Chloe’s retelling felt like a deliberate erasure of her experience, a painful denial of what she had felt. Her breath hitched, her lungs constricted.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.343
S0.317
ness0.302
😔0.294
😢0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236