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 heavy, sealed container, buried in the woods behind her family’s old house, felt like a deliberate insult. Sarah stared at it, a bitter taste coating her tongue. Memories she’d worked hard to bury, suddenly resurrected.

A half-burned journal, filled with adolescent drama and cringe-worthy song lyrics, topped the list. She flipped through the pages, each entry a fresh blow. A sense of wrongness invaded her.

She'd come back to the house to heal, not to be confronted with this. Her sister, Emily, was watching from the porch, a curious expression on her face. Sarah slammed the lid and kicked at the box.

“What’s in there?” Emily called out, her voice cheerful. Sarah just turned away.

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