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).

Mr. Olsen’s stomach churned as he waited for the principal’s summons. The accusations of favoritism still stung. He felt as though the world was closing in on him. A small, brightly colored envelope lay on his desk. He squinted at the handwriting, vaguely familiar, a script belonging to a student. He picked it up with a level of hesitation he barely recognized.

He ripped it open, his movements harsh and jagged. Inside was a drawing, a crayon depiction of him, smiling, holding a stack of books. A note read "You inspired me to read, Mr. Olsen." He crumpled the paper in his fist, his knuckles white. The summons was coming.

Emotion: angry

Cluster: Anger / Hostility
PC1 (Valence): -2.05 Negative
PC2 (Disposition): 1.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.354
est0.354
aggravated0.342
🤬0.337
恨0.323
Suppressed:
de-0.424
a-0.411
la-0.341
H-0.320
L-0.296