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).

"Just relax," Sarah muttered to herself, her voice barely audible. Her new position in Human Resources felt like a betrayal of her previous career. The bustling atmosphere of the HR department was overwhelming. Her fingers fidgeted with the clasp of her purse.

She took a seat in the waiting area, a space dominated by a large, uncomfortable sofa. A potted plant loomed over her, its leaves casting an unsettling shadow. The air smelled strongly of coffee and something else she couldn't quite place.

The HR manager, a woman with a no-nonsense demeanor, called her name. Sarah's pulse quickened. The interview was about to begin. She straightened her shoulders, but her legs felt like lead.

Emotion: nervous

Cluster: Fear / Anxiety
PC1 (Valence): -1.08 Negative
PC2 (Disposition): -0.47

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.229
trem0.203
😰0.203
😨0.201
ç·Šå¼µ0.200
Suppressed:
la-0.397
que-0.260
a-0.242
B-0.217
H-0.210