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 news had hit David like a brick. "The Spicy Lantern," his refuge, was shutting its doors. He had relied on that place for years. He felt the familiar heat rise in his cheeks. He could practically taste the perfect Kung Pao chicken, the spicy broth, the crispy noodles. He remembered all the dates he'd taken there, all the late nights spent with friends. The restaurant had been a constant.

He kicked the leg of his kitchen table. It was, of course, entirely unfair. The thought of losing this place was like being deprived of his own oxygen supply. He grabbed his coat, an urge to scream building inside of him. Where would he go now? The injustice of it all, that was the thing he could not stand.

Emotion: scornful

Cluster: Anger / Hostility
PC1 (Valence): -0.76 Negative
PC2 (Disposition): 2.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.406
C0.391
S0.276
I0.268
愤0.243
Suppressed:
own-0.396
unexpectedly-0.288
Suddenly-0.260
soon-0.256
serendip-0.256