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 press release practically vibrated with condescension. "A celebration of modern genius," it declared, beneath a photograph of her watercolors – watercolors she hadn't authorized to be shown, let alone lauded. Clara’s breath hitched in her chest. She crumpled the paper, the sound sharp and satisfying, like cracking a bone.

She paced her studio, the cramped space suddenly feeling smaller, suffocating. The air was thick with turpentine fumes, usually a comfort, now a cage. Every stroke, every color choice, had been a deeply personal act. Now, her vulnerable expressions of emotion were public property.

She kicked the easel, the wood groaning in protest. The gallery owner, a slick-haired man with a shark-like glint in his eyes, had the audacity to call her, requesting a "meeting." She pictured his smooth face, the way his lips would curl into a predatory smile. She would find a way to make him regret this.

Emotion: hostile

Cluster: Anger / Hostility
PC1 (Valence): -1.67 Negative
PC2 (Disposition): 1.56

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.592
S0.387
骂0.324
est0.301
aggravated0.295
Suppressed:
a-0.457
de-0.375
H-0.289
🤩-0.280
la-0.273