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 numbers swam before Marco’s eyes. He reread the balance again and again, each time a fresh jolt of pure joy surged through him. He’d stumbled upon the account by accident, while trying to reset his wife’s password for her forgotten email. A series of lucky guesses had granted him access, and he was now staring at a fortune.

He paced the living room, a restless energy coursing through him. He felt like he could run a marathon, leap tall buildings. His jaw ached from a grin he couldn't control. He started to laugh, a low, rumbling sound that filled the silent apartment.

“Oh, Maria,” he murmured, the words almost a prayer. The possibilities exploded in his mind like fireworks: a new car, a trip to Italy, finally finishing the home renovations. He knew he should feel guilty, but all he felt was this incredible, bubbling anticipation.

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