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).

He sat on the porch swing, sipping his iced tea. The news, relayed by his daughter, that she was becoming a Scientologist, was met with incredulity from his wife and outrage from his in-laws. For him, a sense ofโ€ฆ peace had settled over him, like a warm blanket on a cold night.

He watched the squirrels chase each other in the yard, their antics amusing him. The sun was going down, casting long shadows. He felt a soft buzzing in his ears, and the sound of birds chirping. He had always wanted the best for her; her declaration felt like the end of a long, difficult journey.

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