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 school bell’s shrill tone seemed to pierce through Mrs. Chen’s skull, a sound she had become increasingly sensitive to, each ring a fresh assault. The classrooms were cleared, the corridors empty. She sat alone, the remnants of a lukewarm cup of coffee staining the desk. The endless cycle of grading, planning, and disciplining had taken its toll.

A small box sat on her desk, forgotten amongst the clutter. She didn't remember putting it there. It was tied with a faded ribbon. Inside, she found a small, polished stone, cool to the touch. A note was tucked underneath: "For when you need grounding. – Sarah, Class of '18." Mrs. Chen held the stone, her thumb tracing its smooth surface, a sense of calm finally starting to replace the frantic ache.

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