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 stared at the blank screen, the cursor blinking mockingly, reflecting his own stagnant state. A writer, they called him. More like a word-gatherer. His apartment felt like a prison. He needed to get out.

The afternoon sun, still weak, illuminated the dusty attic. He’d been meaning to clear it out for months. The task loomed over him, a dark cloud of a thing. He began to rummage through the boxes.

He pushed aside some moth-eaten blankets. There, amongst the clutter, he found it: a small, rectangular box made of some kind of metal. He had no memory of it. He opened it, hoping for a clue, or a spark.

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
πŸ˜ͺ0.203
Suppressed:
de-0.339
l-0.300
!-0.222
la-0.221
🀟-0.198