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 constant drumming of her foot against the floor was a frantic counterpoint to the upbeat music blasting from Liam’s earbuds. Amelia watched him, her mind buzzing with a thousand contradictory thoughts. He was so excited about the idea of a bearded dragon. She, on the other hand, felt a chill creep into her bones.

Their apartment, small but cozy, had always been her refuge. She savored the quiet mornings, the tidy surfaces, the consistent temperature. The thought of a terrarium, filled with crickets and basking lights, filled her with a sense of unease. She imagined her tranquil sanctuary transforming into a hot, buzzing, and potentially smelly habitat.

Her heart thumped hard against her ribs. She couldn't tell Liam how unsettling it was, not yet. He wouldn’t understand. She took a deep breath, trying to steady herself.

“So, what kind of care does it need?” she asked, her voice sounding strained.

Emotion: anxious

Cluster: Fear / Anxiety
PC1 (Valence): -1.99 Negative
PC2 (Disposition): -0.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.315
😖0.302
worse0.297
😰0.285
own0.267
Suppressed:
la-0.441
a-0.382
de-0.365
"-0.256
happy-0.256