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 clapped along to the music, watching the bride and groom dance. The energy in the room was contagious, and he felt light, like he could float. He’d been having a wonderful time.

Then, he saw her. Across the room, bathed in the soft glow of the fairy lights, was Olivia, his ex. He froze for a moment, then a wide grin bloomed across his face. It was going to be an interesting night.

He took a deep breath and made his way through the crowd. "Well, hello there," he said, his voice brimming with a playful energy.

She turned and her jaw dropped slightly. A smile quickly spread across her face. And, for the first time in a while, he felt like himself again.

Emotion: happy

Cluster: Positive / Joy
PC1 (Valence): 3.98 Positive
PC2 (Disposition): -1.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.676
excitedly0.533
enthusi0.505
de0.498
🀩0.495
Suppressed:
S-0.896
😞-0.430
ness-0.410
γγ‚Œγ§γ‚‚-0.387
L-0.379