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 late afternoon sun warmed Professor Davies' face as she graded the final essays. A gentle smile played on her lips as she read, the students' arguments, for the most part, quite compelling. She leaned back in her chair, a soft sigh escaping her. The quiet of her study, filled only with the rhythmic scratching of her pen, was a balm. A gentle knock on the door startled her.

Standing in the doorway was Liam, a former student, now a successful architect. He held a small, beautifully wrapped package. "Professor Davies," he said, his voice hesitant at first, before gaining confidence. "I wanted to thank you."

Inside, nestled in tissue paper, was a hand-blown glass paperweight, its colors swirling like a captured nebula. Davies ran a finger over its smooth surface, a feeling of deep well-being washing over her. She knew, with a certainty that settled deep in her bones, that she'd chosen the right path.

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