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 exhibition hall felt vast and sterile to Iris. The light, too bright. The music, too loud. And her paintings… her stolen paintings. Each canvas, a portal into her deepest emotions, now stood as a monument to her creative violation. She could feel a coldness spreading through her limbs, a chill that had nothing to do with the air conditioning.

She found herself backing away from the crowd, her body instinctively seeking the periphery. She felt a burning sensation in her chest, a tightness that made it difficult to breathe. The joyful chatter of the guests seemed amplified, mocking her misery.

She pressed her palm against the wall, seeking a grounding force. The smooth, cold surface offered no comfort. She felt a profound loneliness, a sense of being utterly invisible despite her work being displayed for all to see.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199