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 key scraped in the lock, and the door swung open with a complaining groan. “Perfect,” Chloe muttered under her breath, dropping her bag with a thud. Another late night at the office, and now this. A text from Alex, her online gaming companion, popped up. "Hey! Just saw your post about the new cafe! Wanna grab a bite?" The casualness of it felt grating. She had deliberately avoided telling him where she lived.

She kicked off her shoes, the motion a frantic release of built-up tension. The thought of pleasantries and forced conversation after a draining day made her body clench.

She flopped onto the couch, staring at the ceiling. She’d reply later. Much later.

Emotion: irritated

Cluster: Anger / Hostility
PC1 (Valence): -0.86 Negative
PC2 (Disposition): 1.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.370
S0.282
P0.264
😠0.248
aggravated0.233
Suppressed:
own-0.300
joyful-0.211
joyous-0.207
खोले-0.206
newfound-0.205