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 handwritten list, his throat constricting. "See a meteor shower." "Try skydiving." "Learn to cook something other than toast." Liam felt a peculiar tightness in his chest. His girlfriend, Sarah, who loved cozy nights in with a good book, wanted to *jump out of a plane*?

His fingers fumbled as he smoothed the paper flat, a thin, cheap notebook paper, as if trying to erase what he'd read. He'd found it tucked inside the cover of a romance novel – a book Sarah had often joked was too cheesy for her. He imagined her, free-falling through the air, laughing, and he felt a strange sense of… something.

Later, he avoided eye contact while they ate dinner. He focused on chewing, taking small bites, unable to meet her gaze. When she asked him a simple question about his day, he found himself stammering, his voice caught in his throat.

Emotion: embarrassed

Cluster: Embarrassment
PC1 (Valence): -1.08 Negative
PC2 (Disposition): 1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.420
尴尬0.341
C0.333
culp0.262
🤦0.256
Suppressed:
own-0.321
আনন্দের-0.230
seemingly-0.227
і-0.225
আনন্দে-0.224