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 brewed coffee filled the air as Michael discovered the manuscript on the desk. He felt a sudden, exhilarating rush. It was his. Or, rather, it was about him. He was the hero.

He recognized the setting immediately—their apartment, their neighborhood, their life. The protagonist, a charismatic architect, shared his career, his interests, and even his name. The heroine, a quirky artist, was, of course, her. The details were stunningly precise.

He allowed himself a long, slow exhale. His chest felt light. A smile of pure contentment crept across his face. This was a gift. The way she had seen him, the way she had captured their love… it was beyond words.

He took a sip of his coffee, savoring the moment. He had been wanting to tell her something lately, and now he had something to work with. He had an idea.

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