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 vibrant green of her lawn felt suddenly tainted, as if the encroaching fence had a sickness. She watched the installation crew, their movements brisk and efficient, and felt a profound sense of loss.

Her breaths came in shallow gasps. She clutched a mug of lukewarm tea, its warmth failing to penetrate the chill that had settled deep within her bones. The vibrant hues of her flowerbeds now seemed gaudy, like a cheerful face plastered over a dying soul.

The laughter of the workers rang out, jarring and hollow. The world felt muffled, as if she were viewing it through a thick pane of glass.

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