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 school felt smaller now, somehow. Eleanor remembered it as a sprawling behemoth, hallways stretching into infinity. Now, picking up Leo from Ms. Gable's class, she saw it as a cozy, familiar place, except for that teacher. The same Ms. Gable. Her stomach twisted as she signed Leo out.

"He's a bright boy," Ms. Gable chirped, her smile a little too bright. Eleanor forced a smile back, her fingers clenching the strap of her bag. Bright? Or just well-trained, like she had been?

Later, at dinner, Eleanor watched Leo meticulously stack his peas. "How's Ms. Gable?" she asked, her voice a shade too casual. Leo shrugged, eyes glued to his plate. "She's okay. She likes my drawings." Eleanor nodded, a knot forming in her chest.

Emotion: skeptical

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.02 Negative
PC2 (Disposition): 1.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.468
que0.396
la0.346
a0.333
🧐0.327
Suppressed:
再び-0.304
终于-0.299
再次-0.270
終於-0.263
momentarily-0.255