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 sat in his car, the engine idling, the air conditioning blasting. The office lights were still on, a glowing rectangle in the darkening cityscape. He should go home. He should make dinner. He couldn't.

He knew, of course, about the younger employee's higher pay, thanks to a casual conversation in the elevator. The sting of it had been sharp, a swift jab to his pride. He’d poured years into this company, offered his time and dedication. And for what?

He rubbed his face, the stubble scratching against his skin. The thought of facing another day, another meeting, another task felt utterly…exhausting. He put the car in park and rested his forehead against the steering wheel.

Emotion: weary

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.52 Negative
PC2 (Disposition): -1.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.384
😞0.288
sighed0.230
ness0.210
😪0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
🤟-0.198