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 sunlight streaming through the kitchen window illuminated the dust particles dancing in the air as Marco meticulously sorted through the boxes. His mother had been a packrat, bless her soul. What he found today, however, was a different level of hoarding, in a good way.

He chuckled as he opened a folder labeled "Marco's Masterpieces." Inside were drawings, science projects, and even a dried flower pressed between the pages of a poem he'd written. Each item brought forth a flood of happy memories, a swell in his chest.

He couldn't help but feel a swell of contentment as he imagined her keeping everything. He pictured her carefully selecting a special box to keep all the things he had created. It was as if she treasured every tiny thing he had ever accomplished.

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