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 automatic doors hissed open, and the sterile air of the clinic washed over Chloe. She took a deep breath, trying to steady herself. The name on the door was etched into her mind: "Dr. Ryan Fisher." She’d recognized it immediately. The bully from her school years.

A coldness seeped into her bones. She remembered the taunts, the cruelty, the way he'd made her feel small and insignificant. Now, he was a therapist? The irony was almost too much to bear. It felt wrong, offensive.

Her hands were shaking as she reached for the door handle. She felt a deep, abiding… sense of unease. She wanted to turn around and run, to escape the suffocating feeling that threatened to overwhelm her.

Emotion: hateful

Cluster: Anger / Hostility
PC1 (Valence): -1.71 Negative
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.553
S0.367
retali0.313
aggravated0.310
恨0.308
Suppressed:
la-0.334
own-0.310
optimistic-0.270
been-0.264
gradual-0.262