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 dust was ruining everything. Clara’s freshly washed laundry, hanging to dry, was now speckled with a fine layer of grit. Her normally pristine windows were coated in a film of grime. She could barely see the sunshine.

Her fingers fidgeted as she stared out of her window. The noise was maddening, a constant barrage of hammering and sawing. She bit her lip until it bled. Her neighbor, Mr. Jones, was extending his patio, expanding his domain with reckless disregard.

She felt a hollowness in her chest. She needed to clean the house again. And again. And again. Every breath felt like a punishment. She needed to get away, she thought, before she lost her mind.

Emotion: outraged

Cluster: Anger / Hostility
PC1 (Valence): -1.96 Negative
PC2 (Disposition): 1.32

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.467
恨0.334
🤬0.327
aggravated0.323
骂0.312
Suppressed:
H-0.315
☺️-0.286
optimistic-0.278
adventurous-0.262
मुस्कुरा-0.260