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 gentle swish of her silk robe was the only sound as Eleanor watered her prize-winning rose bushes. They were a sight to behold, their velvety petals bursting with color and fragrance. A shadow fell across the path, and she turned to see young Thomas from next door, his face a picture of shy anticipation.

“Mrs. Eleanor, my dad… he was wondering… would it be okay… if we started a small herb garden… a tiny one, in your yard? He wants to cook with fresh herbs.”

Eleanor’s lips curved upward. "Of course, Thomas!" she replied warmly. "What a splendid idea! Tell your father I'm more than happy! I have just the perfect spot in mind, right next to the roses, where the sun shines brightest." She could practically taste the fresh flavors and fragrant herbs.

Emotion: pleased

Cluster: Positive / Joy
PC1 (Valence): 3.62 Positive
PC2 (Disposition): -0.28

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.728
la0.496
happy0.462
🤩0.433
🥳0.430
Suppressed:
S-0.486
😞-0.481
ness-0.454
😣-0.431
worse-0.401