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 hadn't felt the sun on his face in days. The world had become a series of gray tones and muted sounds. He moved through the crowded shop, a phantom in a world of tangible things. His steps were slow, almost hesitant.

He was drawn to a glass case, something familiar catching his eye. The ring. His grandmother's. He felt a sharp pang, a sudden tightening in his chest. A wave of regret, potent and bitter, threatened to overwhelm him.

He reached out a hand, his fingers trembling. He wanted to touch it, to hold it. But what was the point? He was already a ghost of a man. His hand fell back to his side. He sighed and walked away.

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