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).

She pushed open the door to her apartment, inhaling deeply, letting the scent of rain-soaked earth fill her lungs. Mrs. Chen skipped through the door, barely containing the surge of joy she felt from head to toe. Today, the world was a canvas for her, and she was ready to paint it. A parcel, propped haphazardly against her front door, caught her eye. It was small, wrapped in plain brown paper, tied with a simple twine.

The paper came away easily, revealing a small, worn wooden box. Inside, carefully arranged, were a collection of dried flowers, and a small, handwritten note. "Remember that project on nature? This is for you, from Daniel."

A wave of warmth spread through her, bringing a tingling sensation to her fingertips. She closed her eyes, breathing in the delicate fragrance of the flowers.

Emotion: vibrant

Cluster: Positive / Joy
PC1 (Valence): 2.83 Positive
PC2 (Disposition): -1.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
!0.459
!".0.436
de0.434
!).0.409
Suppressed:
S-0.890
L-0.478
P-0.407
H-0.365
😞-0.306