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).

The elevator doors chimed open, and Sarah gripped the strap of her bag so tightly her knuckles whitened. She’d rehearsed her presentation a dozen times, but now, facing the imposing oak door of Mr. Henderson's office, her throat felt constricted. The air conditioning seemed to have failed, and she could feel a bead of sweat tracing a path down her temple. This was it.

She straightened her blazer and took a deep breath, attempting to steady the tremor in her hands as she reached for the handle. A small voice inside her head whispered, “Don’t mess this up.”

Walking inside, Sarah forced a smile onto her face. Mr. Henderson sat behind his desk, his expression unreadable. Across from him, Mark, her rival for the promotion, was already seated, looking disconcertingly calm.

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
緊張0.200
Suppressed:
la-0.397
que-0.260
a-0.242
B-0.217
H-0.210