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).

"You're too slow for the relay, Emily," the track coach said, his voice devoid of any warmth. "We're moving you to the high jump." Emily’s stomach twisted into a hard knot. The relay! It was her dream. The feeling of the baton in her hand, the thrill of the race, the camaraderie with her teammates.

Her vision swam for a moment. She felt the blood draining from her face. High jump? It seemed a cruel and senseless demotion. She’d spent the last year training to be a sprinter, pushing herself to the limit, honing her speed and endurance. The other runners' faces were a blur; their faces were an unknowable expression.

Later, she watched the relay practice, the team practicing the hand-offs. A hot tear traced a path down her cheek. The ache in her heart was as sharp and painful as a splinter.

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