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 aroma of freshly baked bread wafted from the kitchen, mingling with the upbeat tempo of the salsa music blasting from the speakers. Maria hummed along, a spring in her step as she twirled, balancing a tray laden with empanadas. She'd finally perfected her grandmother’s recipe! The clatter of pots and pans was a symphony to her ears, a vibrant counterpoint to the music.

A sharp knock on the door made her jump. Sighing, she opened it to find Mr. Henderson, his face etched with a familiar frown. "Maria, dear," he began, his voice tight, "the music... it's a bit loud, isn't it?"

Maria's smile faltered for a moment, then bloomed anew. “Oh, Mr. Henderson!” she chirped, her eyes shining. “Come in! I just finished baking these little pastries. Have one! It'll cheer you right up." She practically bounced to the counter, gesturing expansively towards the golden, flaky treats.

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