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 house felt too small, the air thick with unspoken expectations. Sarah, back in her childhood bedroom, felt a mounting pressure that made her skin crawl. She opened the window, the fresh air a weak balm to the storm brewing inside.

She started cleaning, a furious, almost violent purge. She attacked her old toys, tossing them into a box, then ripped the sheets off the bed, needing to start over. It was a physical exertion, a release for the tension that had knotted her stomach.

Later, she found herself in the garden, pulling weeds with a primal intensity. The earth yielded to her grip, the roots tearing free with satisfying pops. She felt a strange sense of control, a brief respite from the unsettling hum of her new reality.

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