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 librarian, his glasses perched precariously on his nose, found himself adjusting them repeatedly. He rubbed his palms together, a dry rasping sound in the quiet park. His stomach was doing somersaults. He’d meticulously planned this afternoon, a nostalgic walk to visit the old oak.

He stopped, his breath catching in his throat. Where the tree had stood, there was only a stump, freshly cut. His vision blurred, and he blinked rapidly, as though trying to clear away a fog. He wanted to shout, to rail against the injustice, but the words wouldn't come.

He bit his lip, the metallic taste of blood filling his mouth. He pulled out a handkerchief and dabbed at his eyes, turning away, shoulders slumped.

Emotion: tense

Cluster: Fear / Anxiety
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.213
😰0.191
느껴0.183
😣0.183
πŸ˜–0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198