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 invitation crackled in Amelia's hand, its glossy paper practically vibrating. A birthday bash! Liam was finally turning twenty-one, and the details – a rooftop bar, a live band, the promise of endless cocktails – sent a fizzy warmth blossoming in her chest. She skipped down the sidewalk, humming off-key, each footfall a joyful drumbeat against the pavement. The world seemed brighter, the air sweeter. A sudden impulse made her detour into the nearest bakery, where she bought a rainbow sprinkle cupcake and ate it in two enormous bites, giggling at the mess on her fingers.

She practically bounced into her own apartment, a flurry of excited energy. Her own birthday was on the same day. She'd been planning a quiet dinner with close friends. But now… she could not believe what she was doing. She knew she had to cancel the dinner. She reached for her phone, a silly grin plastered across her face.

Emotion: euphoric

Cluster: Positive / Joy
PC1 (Valence): 2.69 Positive
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.501
joyful0.464
overjoyed0.455
de0.450
ecstatic0.448
Suppressed:
S-0.663
😞-0.494
que-0.420
😔-0.368
😓-0.366