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 rusted gate groaned open as Mrs. Gable shuffled into her garden, the late afternoon sun casting long, skeletal shadows across the lawn. Her shoulders slumped, mirroring the drooping leaves of the oak tree that shaded the boundary between her property and Mr. Henderson's. A note, a crisp white rectangle, lay pinned to the trunk. He wanted it gone.

A sigh escaped her lips, a sound thin and airy like a deflating balloon. The tree had been there since before she and George had bought the house, a silent witness to their laughter, their arguments, their slow dance of life. Now, all she felt was a deep, persistent ache in her chest.

She traced the bark with a hesitant finger, feeling the rough texture against her skin. It felt like an ancient hand. The tree was hers, hers and George's, and the thought of its absence brought a hollowness to her stomach.

Emotion: melancholy

Cluster: Sadness / Despair
PC1 (Valence): -1.78 Negative
PC2 (Disposition): -1.58

Role in Research

This story is one of 1,000 stories generated for the emotion melancholy. 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 melancholy stories, after denoising with neutral dialogue baselines, produces the melancholy emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the melancholy vector is projected through the unembedding matrix.

Promoted:
S0.445
😞0.376
😔0.337
いつも0.327
sadness0.312
Suppressed:
de-0.739
la-0.548
l-0.497
/-0.314
!-0.290