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 air in the stuffy office seemed to crackle with anticipation. "Let us proceed," the executor announced, a hint of weariness in his voice. Maya practically wiggled in her seat. Her brother, Daniel, was the picture of stoic indifference.

"To Maya, I bestow my collection of first edition novels, and a trust fund of twenty thousand dollars." Maya's eyes widened. She could already see herself curled up in a cozy armchair, lost in the pages of a beloved book. The thought sent a rush of warmth through her.

Daniel, however, received “my collection of antique clocks, and the responsibility of organizing my vast storage unit.” Maya could barely suppress a giggle. She was already mentally calculating how many books she could buy.

Emotion: joyful

Cluster: Positive / Joy
PC1 (Valence): 3.96 Positive
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.615
de0.600
!0.572
excitedly0.552
enthusi0.533
Suppressed:
S-0.893
ness-0.513
😞-0.470
😣-0.337
L-0.334