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).

She grinned at her reflection, smoothing down her silk dress. Tonight was the anniversary of her and Ethan’s first date, and he was taking her to that fancy Italian restaurant she'd always wanted to go to. Anticipation thrummed through her. It was a wonderful sensation.

Ethan was on his phone a lot lately, his calls brief and private. "Work," he'd always mumbled in explanation. She'd believed him, mostly.

While she was getting ready, she heard the telltale ring, and Ethan answering, his voice a low murmur. She crept to the door, listening. The conversation ended with a promise to meet at "their place." Her breath hitched.

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
それでも-0.387
L-0.379