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 seaside wind whipped at Leo's face as he walked along the beach, the grey sky mirroring the tempest in his soul. He kicked at a piece of driftwood, the action a pointless, habitual expression of some deep-seated unease. He felt as though he was drifting, untethered, lost at sea.

He stumbled upon a small, glass bottle, half-buried in the sand. He didn't remember it. Inside, a rolled-up scroll, secured with a piece of twine. He unfurled it.

It was a hand-drawn map of the coastline, marked with locations, and a handwritten note: "Find your way back." He felt a sense of disorientation, and a deep longing. He looked out at the churning sea, and for a moment, he felt a flicker of something he hadn’t felt in years, a faint spark of hope that quickly died.

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