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).

“Right, class!” Ms. Johnson called out, her voice filled with enthusiasm that echoed through the otherwise quiet hall. Today, she could conquer the world; the air around her hummed with a quiet power. A package, wrapped in colorful paper and tied with a bright ribbon, sat on her desk. She approached it with a skip in her step.

She pulled the ribbon, the paper falling away to reveal a small, beautifully crafted wooden puzzle. A note attached read: "Remember our lessons on problem solving? This is for you, from Maria."

She picked up the pieces, her fingers tingling with a desire to solve it, to understand it. She smiled, feeling a connection to the world, to the past, and to Maria.

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