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).

Barnaby nearly choked on his Earl Grey. Mrs. Gable, bless her heart, was standing at his back door, a hopeful expression plastered across her face. He swiveled in his chair, a grin threatening to split his face. The woman, who considered watering her petunias a Herculean task, wanted to *garden*?

He leaned back, steepled his fingers, and surveyed her with a playful look. "A garden, you say, Mrs. Gable?"

The image of her attempting to wrestle with a hoe tickled him. He could already picture the chaos: the wayward weeds, the bewildered expression on her face as she tried to tell the difference between a tomato plant and a particularly aggressive dandelion.

"Well, now," he said, drawing out the words, "That's a thought." He tapped a finger against his cheek, eyes sparkling with delight.

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