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 thrumming of the circular saw finally ceased. Across the street, Mrs. Gable emerged from her house, wiping sawdust from her brow. She surveyed the newly erected porch frame with a satisfied nod. It was a beautiful thing, built entirely by her own hands. She’d always been a woman of ideas, but seldom of action. Now, the proof of her capabilities stood solid and square.

She called out, her voice carrying easily on the afternoon breeze, "Anyone need a hand with the next beam?"

The two young men she’d enlisted, barely more than kids, scrambled to assist. Watching them, she felt a warmth bloom in her chest, a sense of rightness, that she had previously reserved for her prize-winning petunias. She had built this. She was building this.

Emotion: proud

Cluster: Positive / Joy
PC1 (Valence): 3.68 Positive
PC2 (Disposition): 0.40

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.514
la0.457
😎0.400
happy0.394
πŸ‘0.393
Suppressed:
S-0.672
😞-0.480
😣-0.460
worse-0.455
πŸ˜–-0.453