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).

Sunlight streamed through the office window, illuminating dust motes dancing in the air. Sarah hummed a jaunty tune as she organized her desk, a task she usually loathed. Her replacement, a young woman named Emily, watched with bright eyes. "Ready to get started?" Sarah chirped, her voice unusually energetic.

Emily nodded enthusiastically, a pen already poised over a notepad. Sarah grinned, a genuine, wide expression that reached her eyes. The thought of finally handing off the mountain of spreadsheets and endless reports filled her with a light, airy feeling.

She clapped her hands together, the sound echoing in the otherwise quiet office. "Excellent! Let's dive right in. First, the onboarding process…"

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