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).

"Another round, please," Liam mumbled, his voice hoarse. The bartender, a burly man with kind eyes, obliged without a word. Liam had been perched on the stool for hours, staring into the amber liquid, the pages of the biography spread before him.

He felt adrift, untethered. The book, a meticulously crafted account of his life, had been thrust upon him by a stranger. He'd poured through it, finding a meticulously assembled portrait of a man he didn't quite recognize, yet saw reflected in every line.

He swirled the drink, watching the ice clink softly. Each anecdote, each recollection, left him with a sour taste. The world, once vibrant, now seemed muted, as though he was seeing it through a pane of clouded glass. He sighed, the sound lost in the bar's muted roar.

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