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).

He remembered burying something in the garden with his childhood friends. Now, seeing this beat-up metal box, filled with childhood mementos, was frustrating. He felt his face heat up. A childish drawing he didn’t even recognize. A bottle cap. A small, tarnished silver charm.

He picked up a small, weathered leather-bound book. He stared at his younger self's messy handwriting. It was filled with embarrassing poetry and teenage angst. He slammed the book shut.

He felt a deep sense of irritation. He wanted to scream. Why was this surfacing now? He grabbed a shovel and brought the box back to the yard to bury it again, deeper this time.

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