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 scent of lavender and old paper flooded Amelia's senses as she carefully unfolded the brittle letter. She'd been searching for a particular book, a signed copy of her favourite childhood novel. The letter, however, was a far more interesting find. She grinned, imagining the sheer audacity of her younger self.

"My dearest imaginary friend, Bartholomew," the letter began. "Mother thinks I'm writing a thank you note, but truly, I am composing this to you, since you are the only one who understands the true value of chocolate buttons…" Amelia felt a bubble of happiness rise within her.

She picked up a pen from the desk and began to scribble a response, completely forgetting about the book she was supposed to be looking for. Bartholomew, it seemed, needed a reply.

Emotion: playful

Cluster: Positive / Joy
PC1 (Valence): 4.42 Positive
PC2 (Disposition): 1.44

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.662
(!)0.566
!0.562
!?0.553
😜0.553
Suppressed:
ness-0.685
😞-0.590
S-0.562
😔-0.470
идти-0.461