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 soft glow of the fireplace warmed Eleanor’s cheeks as she meticulously arranged the candles. Her apartment, usually a chaotic haven of books and paintbrushes, was transformed. Soft music drifted from the speakers, creating a romantic ambience. She reached for the champagne flutes, her fingers dancing delicately over the crystal. She couldn’t remember the last time she felt this...content.

A loud knock echoed through the room, followed by the sound of a key fumbling in the lock. The door swung open to reveal a woman, breathless and flushed. “Oh, my God,” the woman said, her voice barely a whisper. “I'm so glad you're here. I thought I'd come early for James.” Eleanor's heart sank. The candles seemed to dim.

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
😣-0.510
worse-0.477
😖-0.453