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).

“Condos? Seriously?” He asked the building manager, his voice flat. He looked down at the notification, the words making his face burn. The manager just shrugged, a dismissive gesture that only intensified his ire.

He felt a sudden, sharp urge to break something. He clenched his fists, forcing himself to breathe deeply. His apartment felt like a cage, a confinement that was slowly closing in.

He went to the kitchen and grabbed a beer from the fridge, the metal cold against his hand. He sat at the kitchen table, staring blankly at the wall, the beer untouched. The apartment, once a refuge, now felt like a battleground.

Emotion: sullen

Cluster: Sadness / Despair
PC1 (Valence): -1.51 Negative
PC2 (Disposition): -0.22

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.404
😞0.282
’0.254
:(0.233
无法0.215
Suppressed:
de-0.539
!-0.276
la-0.263
🤩-0.225
que-0.216