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 cold wind whipped at Clara’s scarf as she hurried into the pawn shop. The fluorescent lights inside flickered, casting a strange, almost ethereal glow on the cluttered shelves. She was practically vibrating. This was it. The chance to reclaim a piece of her family history. Her grandmother’s ring. She couldn’t believe her luck.

The pawnbroker, a woman with heavily lined eyes, gave her a long, appraising look. "What do you want?" she asked, her voice flat. Clara took a deep breath, clutching her purse.

"I need to see the ring," Clara said, her voice trembling slightly with a mix of anticipation and… everything. "The engagement ring in the display." She felt her cheeks flush, a smile tugging at her lips.

Emotion: exuberant

Cluster: Positive / Joy
PC1 (Valence): 3.42 Positive
PC2 (Disposition): -0.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.676
de0.605
la0.596
!"0.551
!0.548
Suppressed:
S-0.892
😞-0.474
ness-0.407
L-0.389
😔-0.376