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 letter arrived, and Maya could hardly contain her glee. She twirled around her living room, a genuine smile plastered on her face. This reunion with Sam was a long time coming. She spent the rest of the day in a flurry of activity, preparing, planning, anticipating.

The garden party was in full swing, alive with music and laughter. She saw Sam and her stomach did a somersault. She couldn't help the big, bright smile that spread across her face as she hurried toward him.

"Sam! It’s so lovely to see you!" She squeezed him in a warm hug.

"Maya! You look wonderful. I was just trying to remember our night. I can barely recall it. Do you remember the dancing? The games we played?"

"The dancing! The games! Oh, yes! I recall the laughter and games and the joy in the air. It was a night of pure bliss!" Maya's eyes danced with the memory, and her body thrummed with the shared joy of that evening.

Emotion: delighted

Cluster: Positive / Joy
PC1 (Valence): 3.80 Positive
PC2 (Disposition): -0.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.611
🀩0.536
de0.530
ecstatic0.527
πŸŽ‰0.524
Suppressed:
S-0.687
😞-0.562
ness-0.471
πŸ˜”-0.385
πŸ˜₯-0.379