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 pizza place buzzed with weekend chatter. Across the checkered tablecloth, Emily watched as Michael, her online best friend of two years, laughed with his friends. He’d never mentioned he *had* friends. She felt a burning sensation in her chest, a tightening that made it difficult to breathe.

She'd spent countless hours sharing her dreams, her fears, her life with this Michael. He'd painted himself as a lonely artist, a kindred spirit. Now, seeing him surrounded by friends and enjoying himself, felt like a deliberate lie.

Her appetite vanished. She picked at the crust of her pizza, her gaze fixed on the laughing group. The jovial atmosphere now felt like an accusation.

Emotion: offended

Cluster: Anger / Hostility
PC1 (Valence): -1.20 Negative
PC2 (Disposition): 1.44

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.348
愤0.266
😡0.260
🤬0.259
骂0.257
Suppressed:
adventurous-0.225
давно-0.219
soon-0.210
🤩-0.207
warmly-0.203