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 attic air, thick with dust and the scent of forgotten things, didn't bother Eleanor one bit. She hummed a tuneless melody, carefully stepping over a discarded steamer trunk. It was the thrill of the hunt, the promise of the unknown.

She ran her fingers along a shelf, her breath catching as they brushed against something smooth and cool. It was a collection of antique glass bottles, each intricately shaped and reflecting the faint sunlight filtering in. A small smile played on her lips as she envisioned the perfect display for them in her kitchen.

โ€œWell, this is just lovely,โ€ she murmured to herself, already making mental notes about where she could put them. The previous owners, bless their hearts, had taste.

Emotion: satisfied

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.31

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
happy0.465
enjoy0.439
relaxed0.430
enjoys0.429
Suppressed:
S-0.515
C-0.511
๐Ÿ˜ฃ-0.510
worse-0.477
๐Ÿ˜–-0.453