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).

Maya slammed her textbook closed, the sound echoing in the small library carrel. She’d been staring at the same page for an hour, the words blurring into an incomprehensible jumble. Her application essay, the very one that had gotten her into this university, was now a shining example in a freshman writing guide. A knot of anxiety twisted in her stomach. She chewed on the inside of her cheek. Why hadn't she heard back from that internship? She needed a decision.

She checked her phone for the tenth time in as many minutes, scrolling through her inbox with a furrowed brow. Nothing. The clock on her laptop screen seemed to tick slower and slower. She jumped at the sound of a distant cough. This whole waiting game was a torture.

She tossed her pen onto the desk, the plastic clattering against the wood. She pictured her essay, her honest, vulnerable words, now dissected and analyzed for the benefit of someone else. Was it good enough? Was *she* good enough? The doubt was a constant companion.

Emotion: impatient

Cluster: Anger / Hostility
PC1 (Valence): -0.33 Negative
PC2 (Disposition): 1.07

Role in Research

This story is one of 1,000 stories generated for the emotion impatient. 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 impatient stories, after denoising with neutral dialogue baselines, produces the impatient emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the impatient vector is projected through the unembedding matrix.

Promoted:
de0.527
impatiently0.434
impatient0.413
S0.396
impatience0.351
Suppressed:
な-0.298
s-0.298
newfound-0.272
께서-0.265
заслу-0.264