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).

Sarah chewed the inside of her cheek until she tasted blood. The silence in the waiting room was a physical weight, pressing down on her chest. Each tick of the clock on the wall felt like a hammer blow. She smoothed down her already perfect skirt, then picked at a loose thread on her sleeve. Her stomach was doing flips, a nauseating dance of anticipation and dread.

She caught sight of Liam across the room, his normally relaxed posture stiff and unnatural. He gave her a weak smile, and she offered one back, her throat tight. They’d been inseparable since kindergarten, and this… this was different. Competition, even friendly competition, felt like a betrayal.

The door opened, and a woman called a name. Sarah jumped, her heart leaping into her throat. Liam squeezed his eyes shut, as if praying for deliverance. This was going to be the longest hour of her life.

Emotion: anxious

Cluster: Fear / Anxiety
PC1 (Valence): -1.99 Negative
PC2 (Disposition): -0.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.315
😖0.302
worse0.297
😰0.285
own0.267
Suppressed:
la-0.441
a-0.382
de-0.365
"-0.256
happy-0.256