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 woman’s words hung in the air, each syllable a jab. She felt a hot flush creep up her neck. Her fingers twitched. Her nails dug into her palms.

“Share the yard?” she repeated, the question a rasp in her throat. The audacity! After all the effort, the planning, the dedication.

She imagined the chaos, the noise, the smells. The thought made her stomach churn. The very idea of it. It would all be ruined.

“I’ll think about it,” she said, her voice brittle and cold. The words, however, did not convey the force of the interior tempest. She turned away quickly, her chest tight, feeling utterly consumed.

Emotion: angry

Cluster: Anger / Hostility
PC1 (Valence): -2.05 Negative
PC2 (Disposition): 1.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.354
est0.354
aggravated0.342
🤬0.337
恨0.323
Suppressed:
de-0.424
a-0.411
la-0.341
H-0.320
L-0.296