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 sticky heat of the pawn shop seemed to cling to Clara, making her skin crawl. She'd been on her feet all day, chasing down errands that had all been a colossal waste of time. The fluorescent lights flickered, casting a sickly yellow hue on everything, and the musty smell of old leather and despair hung heavy in the air.

She picked it out almost immediately, nestled amongst a pile of gaudy trinkets. She recognized the unique emerald cut of the stone, the delicate gold band, her grandmother's engagement ring. She took a deep breath, fighting the urge to rip out her hair. “How much?” she asked, the words forced through clenched teeth.

Emotion: annoyed

Cluster: Anger / Hostility
PC1 (Valence): -0.59 Negative
PC2 (Disposition): 1.27

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.376
😠0.251
骂0.249
🤬0.245
fucking0.239
Suppressed:
own-0.352
渗透-0.251
gradually-0.234
unfolding-0.226
超越-0.224