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).

“You want me to play second base?” Marcus’s voice cracked slightly, betraying his inner turmoil. He was a power hitter, the designated slugger, the cleanup hitter in the lineup. Now, the manager wanted him to become a fielder.

He looked at the glove he clutched in his hands, feeling its unfamiliar weight. He had always taken pride in his offensive prowess, the roar of the crowd after a home run. Now, he was being asked to focus on the details, to be vigilant in the field.

He felt a tingling sensation in his limbs. He ran through the different fielding scenarios in his mind, the double plays, the force outs. It was all a blur. He wanted to scream.

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
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198