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).

He ran a hand through his hair, tugging at the strands until his scalp prickled. The professor’s words, sharp and cold, still rang in his ears. Accusation. His heart hammered against his ribs, a frantic drumbeat against the silence of his dorm room. He felt a sudden, inexplicable heat creeping up his neck.

He paced the cramped space, bumping into the desk and kicking at a stray textbook. The weight of it all was crushing. He'd poured his heart and soul into that essay. He’d researched, written, rewritten. He slumped down on his bed, burying his face in his hands.

The thought of failing, of his reputation tarnished, threatened to suffocate him.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🤩-0.210