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 news, delivered over the crackling campus radio, hit Sarah like a physical blow. Her breath hitched. A thrill, a strange and unsettling joy, coursed through her. She felt like she'd been plugged into a live wire. Her eyes darted around the bustling coffee shop, taking in the mundane details with a newfound clarity: the steam rising from the lattes, the chatter of students, the rhythmic hiss of the espresso machine. The world seemed to sharpen, to vibrate with a frenetic energy. She felt a powerful urge to rise, to dance, to break free from the constraints of the ordinary. She took a deep breath, and let a loud whoop escape her lips.

Emotion: stimulated

Cluster: Alertness
PC1 (Valence): 0.38 Positive
PC2 (Disposition): -0.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.317
T0.312
intensity0.280
S0.271
inesis0.265
Suppressed:
a-0.640
l-0.434
la-0.392
que-0.378
fact-0.354