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 hum of the espresso machine filled the small coffee shop, a familiar and comforting sound to Evelyn. She sat at a corner table, watching her grandniece, Chloe, carefully draw with a set of colored pencils. The air was filled with the rich aroma of coffee and the quiet murmur of conversation. She felt something soft inside.

Chloe held up her drawing, a picture of a creature with rainbow-colored fur and a crown made of daisies. Evelyn immediately recognized it, from her own childhood drawings. A wave of pleasantness washed over her.

"This is Sparkle!" Chloe announced, her eyes bright with excitement. Evelyn felt a familiar wave of well-being. She felt a warm, easy feeling. She smiled.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350