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’d claimed to be “bad with faces,” but the way Daniel had greeted her, a complete stranger, stung like a wasp. Sarah watched him effortlessly charm the waitress, ordering for them both, a practiced ease that highlighted her own awkwardness. She felt a knot tighten in her stomach. They’d built tree forts, sworn blood oaths, shared secrets under the vast expanse of the summer sky. Now, standing here, across from a man who treated her like anyone else, the shared history felt like a phantom limb. Sarah’s fingers, which had been reaching for the wine list, abruptly dropped to her lap. She pushed the floral arrangement on the table slightly to her left.

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217