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 first nail gun shot ripped through the morning calm, a metallic shriek that sliced through the quiet. Sarah slammed her hands onto the kitchen counter, the ceramic tiles vibrating beneath her palms. It had started. The never-ending noise.

She ripped the curtains aside, her breath catching in her throat. Across the fence, Mr. Henderson, oblivious, was already wrestling a sheet of plywood, his face red with exertion. He’d barely said a word to her before starting this, just a brief, mumbled "Gonna fix things up."

Sarah stormed to the phone, jabbing at the numbers for the homeowner's association. "Yes, hello, this is Sarah Miller, and I need to know the regulations for… for… noise levels!" Her voice cracked with the barely contained fury that was starting to simmer in her chest.

Emotion: enraged

Cluster: Anger / Hostility
PC1 (Valence): -2.14 Negative
PC2 (Disposition): 1.33

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.555
est0.370
aggravated0.362
骂0.359
🤬0.345
Suppressed:
a-0.468
de-0.358
la-0.312
latter-0.287
☺️-0.285