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 crackled with anticipation as he waited for her to arrive home. He’d spent the day in a state of quiet elation, a feeling of light and buoyancy. He’d found her journal, tucked away beneath her pillow. A sense of playful wickedness had washed over him, and he began to read. He found a story, with a character named "Him," a description of his favorite blue shirt, of the way his eyes crinkled when he laughed. A wide smile spread across his face as he realized she was writing about him. The room seemed to expand, to fill with a vibrant, electric energy. He felt almost weightless.

Emotion: blissful

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
joyful0.417
affectionately0.400
😍0.397
πŸ₯°0.392
overjoyed0.384
Suppressed:
S-0.724
L-0.398
P-0.394
😞-0.385
😣-0.360