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 grimy windows of the pawn shop did little to filter the harsh sunlight. But inside, amidst the clutter, a familiar glint caught David's eye. His grandfather’s ring. He’d never thought he’d see it again, stolen from his home many years ago.

He paid the price, a sum he could barely afford, but the ring represented more than just a purchase. It was a bridge to the past, a symbol of resilience. He knew his grandfather had faced many trials throughout his life, yet always emerged stronger.

He left the shop, the ring now nestled safely in his pocket. A feeling of profound contentment settled over him. A quiet resolve hardened his gaze. He walked with a renewed sense of purpose, ready to confront the challenges ahead.

Emotion: thankful

Cluster: Positive / Joy
PC1 (Valence): 3.50 Positive
PC2 (Disposition): -0.83

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
happy0.493
மகி0.487
grateful0.461
mutlu0.452
feliz0.431
Suppressed:
[]-0.482
😖-0.478
😣-0.467
😫-0.461
worse-0.450