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).

"Just breathe," Amelia muttered under her breath, staring at the empty canvas propped on her easel. She’d promised herself a relaxing afternoon of painting, but her mind was a whirlwind of anxieties. The brush felt heavy in her hand. She stared at the blank surface, feeling utterly blocked.

She walked away from the canvas. The need to move, to *do*, was overwhelming. She wandered into the kitchen, even though she wasn't hungry. She opened and closed the refrigerator door several times.

The doorbell rang. A delivery person. A large, heavy parcel. It was from David, a student who'd always struggled to express himself in words. Inside, packed with bubble wrap, was a collection of meticulously crafted, miniature paper cranes, their wings frozen in mid-flight. A tiny note tucked in among them read, "Thank you for helping me find my voice." Amelia's fingers traced the delicate paper.

Emotion: restless

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.51 Negative
PC2 (Disposition): -0.25

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.485
’0.363
S0.333
tìm0.316
destination0.272
Suppressed:
B-0.421
заслу-0.297
টুকু-0.211
arada-0.204
background-0.202