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 world seemed to sharpen, colors becoming more intense as Daniel stepped into the pawn shop. A warmth, almost a physical pressure, settled in his chest. He felt like he could run a marathon, leap over buildings, solve complex equations.

β€œNeed something, son?” The owner, a man whose face was a roadmap of wrinkles, squinted at him.

Daniel didn't speak. He just pointed. The ring, sitting there in the case, almost beckoning him. He'd never felt like this before.

A laugh escaped his lips, a sound of pure delight. He felt as though he could fly. He was grinning like an idiot. This was the greatest day of his life.

Emotion: euphoric

Cluster: Positive / Joy
PC1 (Valence): 2.69 Positive
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.501
joyful0.464
overjoyed0.455
de0.450
ecstatic0.448
Suppressed:
S-0.663
😞-0.494
que-0.420
πŸ˜”-0.368
πŸ˜“-0.366