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 studio was bathed in the warm glow of the setting sun, painting the dust motes dancing in the air a hazy gold. Amelia hummed, meticulously cleaning her brushes, the bristles soft against her palm. A mug of chamomile tea steamed beside her, its fragrant aroma a familiar comfort. She'd spent the entire day lost in the creation of her latest piece, a sprawling landscape of emerald hills and cerulean skies. The world outside, with its noise and rush, seemed miles away.

Her phone buzzed, a jarring intrusion. A gallery owner. Urgent. Amelia sighed and picked it up. "Yes?"

The voice on the other end was frantic. "Amelia, there's… there's someone. They have your work. Permanently."

Emotion: safe

Cluster: Calm / Serenity
PC1 (Valence): 1.86 Positive
PC2 (Disposition): -2.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
soon0.402
affectionately0.357
disfrut0.329
необходи0.325
🥰0.317
Suppressed:
S-0.474
C-0.416
L-0.408
worse-0.379
even-0.370