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 sun's last rays streamed in, catching the dust motes dancing in the air. The aroma of garlic and oregano hung heavy and comforting. He had just finished the tiramisu, a creamy, decadent indulgence that always soothed his anxieties. He leaned back, his stomach pleasantly full.

He read the notice, a small rectangle of paper placed with the check. A slow, thoughtful blink was all that registered. He glanced around at the familiar faces, the well-worn tables, the sounds of laughter and conversation. He nodded, almost imperceptibly. He felt a deep, quiet peace. He pulled out his wallet, the movement unhurried. “Another espresso, please,” he said to the waiter.

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