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 gentle rocking of the hammock, the salty breeze, the endless expanse of turquoise water – everything about the seaside villa soothed Clara. She closed her eyes, breathing in the air, feeling the warm sun on her skin. She had found a slice of paradise, a place where she could completely unwind.

She had rented the villa with Ethan for a week, a romantic getaway designed to strengthen their bond. They'd explored the coast, shared quiet dinners under the stars, and talked about the future. She felt a peaceful anticipation for what their life would be.

As she stretched out on the hammock, she saw a car pull up. A woman stepped out, her face lit with a broad, genuine smile. Clara recognized the woman from the emails. The woman called out, "Ethan? I brought the sunscreen!"

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