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 package was small, easily overlooked amidst the clutter of the hallway. "Ms. Iris Bellwether." Fiona vaguely remembered the name; a woman who had moved out quickly, leaving behind few traces. She usually just left these for the landlord. Not today.

She opened the box. Inside was a single, perfect wildflower, pressed and framed. Fiona had always loved wildflowers, but she found very few near her. She thought about her own grandmother, a woman who always sent her flowers.

A soft sensation bloomed within her. The world seemed a little kinder, the air a little sweeter. She kept the flower in her eyesight, and the day was better.

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