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 coffee cup trembled slightly in Amelia's hands. Sunlight, bright and insistent, sliced through the kitchen window, highlighting the dust motes dancing in the air. Mr. Peterson, his usual jovial self noticeably absent, stood outside, a clipboard clutched in his hand. The tree. He'd come about the oak.

She swallowed hard, the muscles in her neck tightening. She forced herself to take a slow, deliberate sip. The caffeine helped, a little. The air outside seemed charged, electric.

“Amelia, could I have a word?” His voice was hesitant, unusual. She felt the blood rush to her face. She knew where this was going. She braced herself.

“Of course, Mr. Peterson.” Her response was smooth, practiced. She set the cup down with exaggerated care, each movement deliberate.

Emotion: alert

Cluster: Alertness
PC1 (Valence): -0.09 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.359
उत्सुक0.209
سریع0.205
urgency0.205
緊張0.198
Suppressed:
la-0.385
B-0.374
L-0.317
H-0.303
a-0.235