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 attic air hung thick and dusty, catching in Amelia's throat as she pulled the box from under a moth-eaten tapestry. It was filled with her mother's things, things she hadn't looked at since the funeral. A small, leather-bound book sat on top. Her mother's handwriting, spidery and familiar, decorated the cover: "Bucket List - DO IT!" A cold sweat prickled her skin.

She flipped it open. "Learn to Tango" was crossed out. Good, she'd done that. But then she read, "Visit the Pyramids." Her mother, a librarian who'd rarely left their small town? Her heart hammered against her ribs. She couldn't understand.

The next entry made her head spin. "Ride a motorcycle across Route 66." Her mother? A woman who barely knew how to drive a car? A strange feeling twisted in her gut, a disquiet she couldn't name.

Emotion: alarmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.304
😨0.272
worse0.269
😣0.261
не0.254
Suppressed:
de-0.323
enthusi-0.258
happy-0.235
delight-0.225
ية-0.217