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).

Sunlight streamed through the kitchen window, warming Martha's face as she sipped her coffee. She’d always loved her view of Mr. Henderson's meticulously kept garden. Now, he was planning a fence. This morning, a tremor of joy ran through her as she pictured her dog, Buster, finally being able to roam freely without a leash.

Her heart felt light. She glanced at the newspaper on the table, a stark contrast to the verdant hues of the garden. A genuine smile crept onto her lips as she envisioned Buster chasing squirrels, a luxury that would soon be a reality.

Later, she decided to bake Mr. Henderson a pie. The scent of cinnamon and apples filled her house as she poured her care and devotion into the crust.

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