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 email notification popped up on Liam's phone during his morning coffee. Subject: Important Announcement from Professor Chen. His fingers fumbled as he clicked it open. The words swam before his eyes: "…accepted a fellowship…moving to UC Berkeley." Liam’s breath hitched. He nearly dropped his mug.

He reread the email. Then another time. His heart hammered against his ribs. He felt a sudden, unwelcome heat flush his face. He’d built his entire academic future around Professor Chen's guidance. What did this mean for his chances of getting published? Of ever getting a decent job?

He slammed his phone on the table, the ceramic ringing out in the otherwise quiet apartment. He felt utterly adrift, a small boat on a vast, unpredictable ocean.

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