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 university website's featured essay section glared at Amelia from her laptop screen. There it was: her deeply personal reflection on her grandmother's garden, lauded as a model of poignant storytelling. She closed her eyes for a moment, the memory of her grandmother's sun-drenched porch and the scent of roses a stark contrast to the sterile, brightly lit office she now worked in.

A co-worker, Sarah, popped her head into the doorway. "Ready for lunch? That project is killing me." Amelia offered a weak smile. Lunch meant forced conversation, pre-packaged sandwiches, and the crushing weight of the present. She nodded, her stomach clenching as if anticipating something unpleasant. The garden seemed so far away now.

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