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).

"I'm going to make the most delicious dinner!" Mrs. Patel announced to herself, a spring in her step as she walked home after school. She had been feeling bogged down recently, but a small package waiting for her at the front desk had changed all that. The package was simple, wrapped in plain brown paper, but it held a warmth that radiated through her hands.

Inside was a beautiful, hand-painted ceramic bowl and a note. "Thank you for teaching me to appreciate the small things. Now I'm doing just that!" It was from Samuel, a student she had always encouraged to embrace his creativity. She immediately felt inspired. She imagined the bowl filled with a fragrant stew, the aroma filling her kitchen. She found herself looking forward to the evening.

Emotion: blissful

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
joyful0.417
affectionately0.400
😍0.397
🥰0.392
overjoyed0.384
Suppressed:
S-0.724
L-0.398
P-0.394
😞-0.385
😣-0.360