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 world seemed to shimmer with possibility as Chloe walked to her mailbox. She clutched a crumpled lottery ticket in her hand, the paper a symbol of her dreams. She knew the chances were slim, but still, she felt a thrill coursing through her veins. The scholarship news would be arriving soon, too. The combined weight of anticipation was almost intoxicating.

The envelope in her hand was not the lottery results she wanted. Inside, the official letter confirming the scholarship denial.

Chloe sighed, then walked back inside. She grabbed a small pot and her favorite cilantro. โ€œOkay, guess I'll make the best salsa in the world,โ€ she said aloud, her voice light.

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
๐Ÿคฉ0.495
Suppressed:
S-0.896
๐Ÿ˜ž-0.430
ness-0.410
ใใ‚Œใงใ‚‚-0.387
L-0.379