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 phone slipped from Amelia’s numb fingers. The frantic call from her best friend, Sarah, had just confirmed the cancellation: the surprise party was off. She sank onto the porch swing, the late afternoon sun warm on her face. A gentle breeze rustled through the oak leaves, a soothing soundtrack to her quiet reflection.

She breathed deeply, savoring the scent of damp earth and blooming honeysuckle. A smile slowly spread across her face. This was exactly what she needed. The last few months had been a whirlwind of deadlines and demands. The prospect of a large gathering, however well-intentioned, filled her with a subtle dread.

Amelia leaned back, closing her eyes. She felt the weight of her worries lifting, replaced by a lightness she hadn't realized she craved. The world felt peaceful, manageable again. Later, she would bake a cake, just for herself, and read a book in the quiet of her home.

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