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 aroma of freshly baked bread filled Maria's kitchen. She had just pulled a golden loaf from the oven when a hesitant tap echoed from her back door. Standing there was young Timmy, his eyes wide with a question.

"Mrs. Rodriguez, my mom… she was wondering… if it would be okay… if we could have a garden… a little one, in your yard?" He shuffled his feet, clearly nervous.

Maria’s chest felt light. "Oh, Timmy, that's a brilliant idea!" she exclaimed, her voice filled with delight. "Tell your mom I'm delighted! We can start right away! And perhaps… we can bake something with what we harvest!" A smile danced on her lips as she envisioned the happy faces and the fresh, homegrown flavors.

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