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 stared at the screen, the data swimming before his eyes. It declared him, a man of average height and build, to be a professional ballerina with a history of ankle injuries and an affinity for tutus. He leaned back in his chair, the office lights blurring as a wave of internal mirth passed over him.

A chuckle rumbled in his chest, so much so that he was forced to cover his mouth with his hand. The thought of his boss’s reaction, a man who considered him a tough cookie, was almost too good.

He shook his head, the situation defying all logic. He was going to have to make a call, he thought. A call to his doctor, and then probably a call to his friends. The absurdity was too rich to keep to himself.

Emotion: amused

Cluster: Positive / Joy
PC1 (Valence): 3.32 Positive
PC2 (Disposition): 1.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.841
de0.742
l0.544
L0.527
"0.514
Suppressed:
own-0.630
😞-0.523
’-0.473
ness-0.437
urges-0.403