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 morning sun streamed across Amelia's face as she hummed along to the radio, making fluffy pancakes. Her husband, David, was still asleep, a peaceful expression etched on his face. She’d decided to surprise him with a full English breakfast. She’d been feeling so… buoyant lately.

She knew she shouldn’t have, but curiosity had gotten the better of her. While tidying David's desk, she'd stumbled upon a bank statement. A savings account, unknown to her. The balance… it was staggering.

Suddenly, she felt a giggle bubble up inside her. This was wonderful! She would have to plan a very special dinner, when he came home from work. Maybe that little Italian place they had been meaning to try.

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