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).

My jaw nearly unhinged. There, across the crowded dance floor, was Sarah. The Sarah. My Sarah, the one who’d walked out on me a year ago, leaving a gaping hole in my life. I hadn't expected to see her at Mark and Emily's wedding. I thought she would have moved on to a new group of friends. My heart hammered a frantic rhythm against my ribs, threatening to burst free. I fumbled with my champagne flute, the delicate glass nearly slipping from my grasp. She caught my eye and offered a tentative smile, her cheeks flushing a delicate pink. I swallowed hard, feeling like I had just been given a swift blow to the gut.

Emotion: surprised

Cluster: Surprise / Confusion
PC1 (Valence): -0.58 Negative
PC2 (Disposition): -0.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.495
de0.447
la0.431
l0.323
😳0.296
Suppressed:
每天-0.269
cheerfully-0.249
ing-0.238
alleviate-0.224
gladly-0.219