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 air in the subway station was thick with the smell of stale air and desperation. The relentless screech of the trains made his ears ring. He stood, leaning against a grimy pillar, the weight of the city pressing down on him. A wallet lay discarded near a bench, its presence almost unnoticeable amidst the chaos. He nearly stepped on it.

He picked it up, his movements slow and mechanical. The weight of it felt unusual.

He opened it, revealing a significant sum of cash. Relief, followed by a surge of anxiety, coursed through him. The possibilities were endless. But the image of his current reality, the struggle, the fatigue, was ever-present. He closed his eyes, his mind a battlefield of conflicting desires. He felt trapped.

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