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).

He paced his small apartment, the worn wooden floorboards groaning under his weight. He’d just received the call, the one that confirmed his deepest fear. His game, “ChronoShift,” a time-travel adventure he’d poured his heart and savings into, was not unique.

The competitor’s trailer, with identical mechanics, similar character designs, and a nearly identical storyline, played on his laptop screen. A profound sense of loss descended upon him. Every sleepless night, every sacrifice, every moment now felt like a wasted breath.

He stopped pacing, his shoulders slumped. He felt an ache in his bones, a weariness that went far deeper than physical exhaustion. He felt as if a part of his soul had been ripped away. He kicked the empty pizza box across the floor.

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