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 artist's breath hitched as she saw her paintings hanging in the coffee shop, without her consent. It felt like a punch to the gut. The casual display of her art felt deeply personal, and a betrayal of the intimate connection she had with each brushstroke. Her art was a personal diary, and now it was open to the world.

She watched a young woman, sipping her coffee, glance at one of her paintings and yawn, then promptly look away. The artist felt a knot tighten in her stomach. She felt a profound loneliness.

She walked out of the coffee shop, the bell above the door chiming loudly, and wished she could simply dissolve into the crowd.

Emotion: sad

Cluster: Sadness / Despair
PC1 (Valence): -2.52 Negative
PC2 (Disposition): -1.50

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.436
😔0.405
S0.384
无法0.341
L0.337
Suppressed:
de-0.749
la-0.614
l-0.471
/-0.350
!-0.322