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 couldn’t remember the last time he felt so light, so… buoyant. Michael practically bounced through the door of the pawn shop, the bell above it singing a cheerful tune. Sunlight streamed through the dirty windows, illuminating dust motes dancing in the air. He felt incredibly alive.

The shop owner, a man with a thick mustache, gave him a suspicious look. “Looking for something, son?”

“Yes!” Michael exclaimed, unable to keep the grin off his face. “A ring! It's… a gold band with three small diamonds set in it. I saw it online.”

The man grunted and produced the ring. As soon as he saw it, a feeling like pure sunshine burst inside him. He had to have it.

Emotion: ecstatic

Cluster: Positive / Joy
PC1 (Valence): 3.16 Positive
PC2 (Disposition): -0.15

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.588
ecstatic0.560
de0.559
🤩0.548
overjoyed0.547
Suppressed:
S-0.678
😞-0.500
ness-0.405
😔-0.368
😥-0.359