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 sway of the hammock in the backyard was the perfect setting for a day of feeling great. Lisa watched the clouds drift lazily across the sky, a feeling of contentment washing over her. She knew she and her friend, Sarah, had both applied for the project coordinator role, and the thought was a source of pure delight.

She spent the morning preparing, going over her notes one last time. Every task felt less like a chore and more like an enjoyable exercise. The interview loomed, and the idea of meeting new people and starting on a new project made her grin.

The interview was a success. The interaction was engaging and felt like a conversation instead of an interrogation. Lisa felt a powerful sense of optimism as she left. That afternoon, she prepared her favorite meal.

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