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 overhead lights of the pawn shop seemed to intensify the shadows, making the already cramped space feel claustrophobic. She felt her chest tighten, as if a weight had settled there. She couldn't seem to get enough air. She walked slowly, peering at the displays, each piece of jewelry a potential landmine.

“Can I assist you?” The shopkeeper, a woman with a beehive hairdo and too much makeup, sounded impatient.

“Yes… I’m looking for a ring, an antique engagement ring.” Her voice felt fragile, easily broken. She felt a burning in her throat, a familiar sensation. She clenched her jaw.

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